ScienceAsia documentation: scias latex.pdf 1

Preparing a manuscript for ScienceAsia using LATEX[this version: 9 Mar 2018; for the latest version see www.scienceasia.org/scias latex.pdf]

Michael A. Allena, Alejandro Saiza, Alfredo Villaroelb, Khamron Mekchayc, Sujin Khomrutaic,Worachart Sirawarapornd,∗

a Physics Department, Mahidol University, Rama 6 Road, Bangkok 10400 Thailandb Faculty of Medicine, Srinakharinwirot University, Sukhumwit 23, Wattana, Bangkok 10110 Thailandc Department of Mathematics and Computer Science, Faculty of Science, Chulalongkorn University,

Bangkok 10330 Thailandd Centre of Excellence for Vectors and Vector-Borne Diseases, 2nd Floor, Science Building 2,

Faculty of Science, Mahidol University, Salaya Campus, 999 Phuttamonthon 4 Road,Nakhon Pathom 73170 Thailand

∗Corresponding author, e-mail: editor@scienceasia.org

ABSTRACT: We summarize the commands and best practices for producing a manuscript written using LATEX forsubmission to ScienceAsia. Most of the details given here also apply to the preparation of any technical document usingLATEX.

KEYWORDS: manuscript preparation, TEX, BIBTEX

CONTENTS

Introduction 1Colour coding of LATEX commands . . . . . 1Getting started . . . . . . . . . . . . . . . . 1

Front and back matter 1Title . . . . . . . . . . . . . . . . . . . . . 1Authors and their affiliations . . . . . . . . 1E-mail address . . . . . . . . . . . . . . . . 2Abstract . . . . . . . . . . . . . . . . . . . 2Keywords . . . . . . . . . . . . . . . . . . 2MSC2010 . . . . . . . . . . . . . . . . . . 3Acknowledgements . . . . . . . . . . . . . 3

Sections 3Headings . . . . . . . . . . . . . . . . . . 3Paragraphs . . . . . . . . . . . . . . . . . . 3Gaps . . . . . . . . . . . . . . . . . . . . . 3Labelling . . . . . . . . . . . . . . . . . . 3

References and citations 3BIBTEX . . . . . . . . . . . . . . . . . . . 3References using BIBTEX . . . . . . . . . . 4References without using BIBTEX . . . . . 4Symbols for non-English languages . . . . 4Citations . . . . . . . . . . . . . . . . . . . 4Links to webpages . . . . . . . . . . . . . . 4

Mathematical expressions 5Subscripts and superscripts . . . . . . . . . 5

To display or not to display . . . . . . . . . 5Functions . . . . . . . . . . . . . . . . . . 5Symbols . . . . . . . . . . . . . . . . . . . 6Accents . . . . . . . . . . . . . . . . . . . 7Space and text . . . . . . . . . . . . . . . . 7Fractions . . . . . . . . . . . . . . . . . . . 8Brackets and other delimiters . . . . . . . . 8Derivatives and integrals . . . . . . . . . . 9Sums, unions, and products . . . . . . . . . 10Matrices and other arrays of quantities . . . 10Piecewise functions . . . . . . . . . . . . . 11Vectors . . . . . . . . . . . . . . . . . . . . 11Equations . . . . . . . . . . . . . . . . . . 12

Chemistry expressions 14Formulae . . . . . . . . . . . . . . . . . . 14Reactions . . . . . . . . . . . . . . . . . . 15Symbols . . . . . . . . . . . . . . . . . . . 15

Float environments 15Tables . . . . . . . . . . . . . . . . . . . . 15Figures . . . . . . . . . . . . . . . . . . . 16Program listings . . . . . . . . . . . . . . . 16

Theorems and proofs 17Theorems and similar environments . . . . 17Proofs . . . . . . . . . . . . . . . . . . . . 18Algorithms . . . . . . . . . . . . . . . . . 18

Fonts and text symbols 18

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 1

Short cuts 18Short versions of commands . . . . . . . . 18Defining your own commands . . . . . . . 18Arguments of commands . . . . . . . . . . 19

Commands not to use 19

Indicating changes in the revised manuscript 19

Submission to other journals 19

Other class files 20

INTRODUCTION

First of all, congratulations on choosing to useLATEX1 with the ScienceAsia class file to prepare yourmanuscript! The final versions of ScienceAsia articlesare typeset using LATEX and so, if your manuscript isaccepted and you follow the guidelines given here, itwill be processed more quickly. Also, manuscriptsprepared in this way will automatically have the cor-rect format and will look good. This will help to createa favourable initial impression with the referees. Ifyou use \cite , \label , and \ref thenthe cross references in your manuscript will havehyperlinks which will make navigating it easier for thereferees.

This document is a detailed guide to the smallnumber of non-standard commands associated withthe ScienceAsia class file and a reference for allthe LATEX tricks you are likely to need. The firstsection deals with commands for the front matter andacknowledgements. The remaining sections covervarious tasks in LATEX in increasing detail. If youare fairly new to LATEX then at least the start of mostof these sections will be essential reading. Even ifyou have some experience with LATEX it is still worthlooking at the sections relevant to you to check thatyou are using the most efficient approach.

This guide should be read in conjunction with theeditorial guide to publishing with ScienceAsia (seewww.scienceasia.org/scias guide.pdf) which dealswith what you should write. The present guide tellsyou how to write it.

Colour coding of LATEX commands

When referred to in the text, standard LATEX com-mands (including those defined in the amsmath pack-age) are shown in black, LATEX commands whichare defined in other standard packages (i.e., thoseavailable from ctan.tug.org) are shown in magenta,standard LATEX commands whose action has beenredefined in a significant way in scias.cls are

shown in red, non-standard LATEX commands (whosedefinitions are in scias.cls) are shown in blue, andexamples of LATEX commands defined for the purposeof this guide are shown in green.

Getting started

If you have not used LATEX before on your computeryou will probably need to install it if you are usingWindows or Mac OS. The software is free. Peoplewith the Windows and Mac OS X operating systemsnormally use the MiKTeX and MacTeX distributions,respectively. LATEX is usually present by default withLinux. If not, or if compiling the ScienceAsia templatefile gives errors, add a basic LATEX installation andthe recommended fonts to your system. E.g., inUbuntu Linux install the texlive-latex-base, texlive-latex-recommended, and texlive-fonts-recommendedpackages and their dependencies.

To create a manuscript for ScienceAsiayou will need to download the zip filewww.scienceasia.org/scias latex.zip which containsthe ScienceAsia class file (scias.cls), theScienceAsia BIBTEX style file (scias.bst), otherstyle (.sty) files which are needed, and the templatefile (scias template.tex). After unpackingthe zip file, place its contents in the same directory(folder) as your manuscript .tex file (or somewhereelse where latex can find it).

Copy the file scias template.tex to a nameof your choice and open it with your favourite text or.tex file editor. Running latex on this newly created.tex file should give no errors (apart from a warningabout an empty bibliography) and the output PDF fileit should produce is shown in Fig. 1.

All characters after % on a line are comments, i.e.,they are ignored by latex (if the % is not immediatelypreceded by an odd number of \ in which case it willappear as the percent sign). When trying somethingout it is good practice to ‘comment out’ the old versionby adding a % in front. If you change your mind aboutthe new version the old one is then easily restored.

FRONT AND BACK MATTER

Title

The title of the manuscript is specified using\title . Remember that only the first word andthe first word after a colon need to be capitalized. Theremaining words should only be capitalized if they areproper nouns.

Authors and their affiliations

Each author should be placed in a separate\author command. The surname should follow

www.scienceasia.org

2 ScienceAsia documentation: scias latex.pdf

Submitted toScienceAsia 1

1

e-mail: ???2

ABSTRACT :3

KEYWORDS :4

INTRODUCTION

Acknowledgements:5

REFERENCES

www.scienceasia.org

Fig. 1 Output PDF file from running latex on a copy ofscias template.tex.

the given name and initials. If a surname is made up ofmore than one word (e.g., El Naschie), then we needto know so that the author’s entry is given correctlyin the annual index of authors. The simplest way toinform us is to replace each space in the surname by anon-breaking space (˜). E.g.,

\authorMohammed S. El˜Naschie

The affiliation(s) of the author(s) are given usingthe \address command. Use a separate com-mand for each address. Do not attempt to format theaddress (by using, e.g., \\).

If authors belong to different affiliations thenthe lower case letters (separated by commas) areplaced in the optional arguments of \author and\address . E.g.,

\author[a]Amita Ramanujan\author[a,b]Charles A. Eppes\author[c,*]Sheldon L. Cooper\address[a]Mathematics Department,California Institute of Scienceand Technology, Oxnard, CA 93030 USA\address[b]Area 51, Edwards Airforce Base,Groom Lake, NV 89320 USA

\address[c]California Institute ofTechnology, High Energy Physics,MC 256-48, Pasadena, CA 91125 USA

The lower case letters are not necessary if all theauthors have the same affiliation(s). E.g.,

\author[*]Rajesh R. Koothrappali\authorHoward J. Wolowitz\addressDepartment of Applied Physics,California Institute of Technology,MC 128-95, Pasadena, CA 91125 USA\addressNASA Goddard Institute for SpaceStudies, New York, NY 10025 USA

Remember that the affiliation is a place of work of theauthor, and not the address of a funding body.

If there is more than one author, a * is placedin the optional argument of the corresponding author,after the lower case letter if present.

E-mail address

The (long-term) e-mail address of the correspondingauthor should be given using \ead . If the corre-sponding author wishes to give two e-mail addresses,the second can be given by placing it in an additional\ead . Note that you do not need to precedeunderscores ( ) in the e-mail address by \ in this case.E.g.,

\eadsheldon_cooper@caltech.edu\eadsheldon.l.cooper@gmail.com

Abstract

The abstract is placed in the argument of\abstract . It should only be one paragraph(and therefore not contain any blank lines). It shouldnot contain any displayed equations (i.e., no \[, \],etc.). You should also not use \displaystyle.Mathematical expressions inside $ $ are permittedbut they should be kept simple as the abstract willneed to be rendered in HTML (without images) forthe ScienceAsia website.

Note that you do not need to write the wordABSTRACT as this is generated automatically.

Keywords

The keywords are given as a comma separatedlist of uncapitalized words or phrases placed inside\keywords . The keywords should all differfrom words in the title.

The keywords for the annual index are selectedlater by the editors, although you are welcome tomake suggestions for this (by placing each entry ina separate \kwidx command) based on the typeof entries in last year’s index.

Note that you do not need to write the wordKEYWORDS as this is generated automatically.

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 3

MSC2010

For mathematics papers, between one and five 2010Mathematics Subject Classification (MSC2010) codesshould be given as a list inside the \MSC com-mand. They should be placed in order of relevanceto your article. See www.ams.org/msc/msc2010.htmlfor further details.

Acknowledgements

If you have them, acknowledgements are put inside\acknowledgements which is placed just be-fore the references. If there are no acknowledgementsthen delete or comment out this command.

SECTIONS

Headings

Section and subsection headings are done using the\section and \subsection commands.A section heading automatically appears in capitalletters – there is no need to write the argument of\section in capital letters. E.g.,

\sectionIntroduction

will appear as in Fig. 1. Any part of a sectionheading which should not appear as capitals shouldbe enclosed in \NoCaseChange . This wouldbe needed for an abbreviation containing lower caseletters but not for lower case letters appearing inside$ $ or \ce . E.g.,

\sectionSolutions of the \NoCaseChangegNLSequation when $p=1/2$

It is also necessary when the heading contains areference to a label containing a lower case letter. E.g.,

\sectionProof ofTheorem˜\NoCaseChange\refT:main

Paragraphs

ScienceAsia follows the default LATEX style of notindenting the paragraph immediately after a heading.You should not try to change this by using \indent.

A new paragraph is made by simply leavingan empty line. Never use \newline or \\ and\indent to get a new paragraph.

Gaps

Sometimes there will be a large gap between blocks oftext and headings or equations. It is often caused bylong blocks of equations which by default LATEX willnot split. You can allow the equations to be split be-tween columns or pages by enclosing the whole blockof equations within \allowpagebreaks . Youshould only be concerned about gaps if you think the

referees will find them distracting. Do not attempt tofix them by using \vspace .

Sometimes there will be large gaps in the bodyof the text where mathematical expressions have beenstretched to ensure the text is right justified. Don’tworry too much about this. If you think it lookstoo ugly, you can try displaying some of the longerexpressions (by placing them inside \[ and \]). Don’tuse \\ within the text.

Labelling

As there are no section or subsection numbers inScienceAsia articles, you should not label sectionsor subsections (using \label ) and then refer tothem (using \ref ).

REFERENCES AND CITATIONS

BIBTEX

If your research group regularly uses LATEX to writearticles we recommend building a BIBTEX database(i.e., one or more .bib files) containing referencesyou cite. Most journals that encourage submission inLATEX supply a bibliography style (.bst) file as wellas a class file. Running bibtex on your .tex file willautomatically produce a list of references in the formatrequired by the journal.

An example of an entry in a .bib file is@ArticleKPR98,author = R A Kraenkel and J G Pereira

and de Rey Neto, E C,title = Linearizability of the

perturbed Burgers equation,journal = PRE,year = 1998,volume = 58,number = 2,pages = 2526-2530,doi = 10.1103/PhysRevE.58.2526

Notice that names are separated by and. The finalword in a name is assumed to be the surname un-less there is a comma in which case the surname istaken as the word(s) before the comma. In the title,words whose case should never be changed (such asproper nouns and abbreviations) should be enclosed inbraces. We suggest that instead of writing the journalname in full or as the standard abbreviation, youinstead write it as a one-word abbreviation (of yourchoice) and give the full name and abbreviation in filescalled, for example, long.bib and short.bib,respectively. For PRE you would have@StringPRE = Physical Review E

in long.bib and@StringPRE = Phys. Rev. E

in short.bib.

www.scienceasia.org

4 ScienceAsia documentation: scias latex.pdf

References using BIBTEX

The line with \bibliographystylesciasshould be uncommented and put somewhere before\begindocument. Replace the begin and endbibliography commands and everything in between by\bibliography . The argument of this com-mand should contain a comma-separated list (withoutspaces) of .bib files with entries for all the citationsin your manuscript with files giving definitions ofabbreviations listed first. E.g., if the entries for thearticles you refer to are in papers.bib then use

\bibliographyshort,papers

if the journal expects abbreviated journal names.When the journal requests your .tex file you shouldreplace \bibliography by the contents of the.bbl file which bibtex has created.

Note that if a field called DOI (which contains thecorrect DOI) is present in a BIBTEX entry, a hyperlinkto the article will appear in the list of references.

References without using BIBTEX

For each reference you cite in the manuscript thereshould be a corresponding \bibitem whose ar-gument is the citation key. This is followed by theactual reference. These \bibitem commandsare placed inside the \thebibliography environ-ment. E.g.,

\beginthebibliography2\bibitemLam=94Lamport L (1994) \textit\LaTeX: A DocumentPreparation System, 2nd edn,Addison-Wesley, Reading, MA.

\bibitemBH04Brihaye Y, Hartmann B (2004)Fullerenic solitons. \textitJ Phys A\textbf37, 1181--1192.\endthebibliography

If you know the DOI of the reference then you maycreate a hyperlink to it using \bihl as in thefollowing example.

\bibitemBH04\bihl10.1088/0305-4470/37/4/006Brihaye Y, Hartmann B (2004)Fullerenic solitons. \textitJ Phys A\textbf37, 1181--1192.

Symbols for non-English languages

For words or names from languages other thanEnglish, accented characters or other text symbolsare sometimes needed. The accent commands(\‘ , \’ , \" , \ˆ , \˜ , \c ,\H , \= , \. , \u , \v , \r ,\t , \d , \b , \k ) can be appliedto any character for which there is a need. If

an accent is applied to an ‘i’ then you shoulduse the dotless ‘i’ obtained using \i. E.g.,Garc\’\ia gives Garcıa. Some examples:Amp\‘ere gives Ampere, Poincar\’e gives Poincare,Schr\"odinger gives Schrodinger, l’H\ˆopital

gives l’Hopital, Do\˜nana gives Donana, gar\ccongives garcon, E\Hotv\Hos gives Eotvos,\=o\.o\uo\vo\ru\too gives oooouoo,\do\bo\ke gives o.o

¯ e. Note that these accent

commands should not be used in mathematicalexpressions.

You might sometimes need the following symbolsas well: \oe, \OE, \ae, \AE, \aa, \AA, \o, \O,\l, \L, \ss. Note that these commands shouldbe enclosed in braces. E.g., \oeuvre gives œu-vre, \OE\ae\AE gives ŒæÆ, T\aakern givesTakern, \AA gives A, Fr\oyland gives Frøyland,\O gives Ø, Jarno\ltowek gives Jarnołtowek,\L gives Ł, Schlo\ss gives Schloß.

Citations

To cite a reference without explicitly referring to ituse \cite where the argument is the key of thereference given either in the entry in one of your .bibfiles (if you are using BIBTEX) or as the argument ofone of the \bibitem commands at the end ofyour manuscript (if you are not using BIBTEX). E.g.,

congratulations on choosing to use\LaTeX\citeLam=94 with the

If you want to refer to a reference explicitly then placethe key inside \refcite instead. E.g.,

See \refciteLam=94 for further details.

gives See Ref. 1 for further details. For more than onecitation at the same time, give a comma-separated listof keys. Never use dashes (i.e., don’t do somethinglike \refciteLam=94--\refciteSlo90) to indi-cate a range of references. The key to each referenceyou want to cite must be given and latex will put adash between three or more neighbouring referencesautomatically. E.g.,

has been done\citeLam=94,BH04.See \refciteLam=94,BH04,Slo90.

gives has been done1, 2. See Refs. 1–3.As a result of a quirk in the production of the

file for the full printed issue we have not managed toresolve, the first citation in your article must be doneusing \cite rather than \refcite .

Links to webpages

If you wish to cite a webpage, it is normally better todo so in the text rather than in the list of references.

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 5

This can be done by placing the URL in the argumentof \url . The text showing the URL is then alsoa hyperlink to the page. E.g., \urlctan.tug.orggives ctan.tug.org. Characters such as in the URL donot need \ before them.

MATHEMATICAL EXPRESSIONS

Mathematical expressions can appear in the textby enclosing them in $ $ or they can be ‘dis-played’ outside of the text by putting the expres-sion in between \[ and \] or inside an equa-tion environment (see later). [Never use $$ $$ or\begincenter $ $ \endcenter.] Note that onlymathematical expressions should appear inside $ $.The correct ways to obtain italics, long dashes, oraccents are given elsewhere in this guide.

Subscripts and superscripts

A single character inside an expression can be madeinto a subscript or superscript by preceding it witha or ˆ, respectively. E.g., $a_ibˆi$ gives aibi. Ifmore than one character is in the sub- or superscript,the characters must be enclosed in braces. E.g.,cm$ˆ-1$ gives cm−1.

It is not necessary to superscript a ‘prime dash’ asit is already treated as being superscripted. E.g., $y’$gives y′. This means that if you want to add anothersuperscript to it you need to include an \mbox

to avoid getting a double superscript error. E.g.,$y’\mboxˆ*$ gives y′∗. Also, $yˆ\prime *$

gives y′∗.If the sub- or superscript causes the overall ex-

pression to have a large enough vertical extent, thespacing between lines will be increased wherever theexpression occurs in the text. To prevent a changein line spacing, the expression should be enclosedin \smash[t] if the superscript is too largeand \smash[b] if the subscript is too large.E.g., $\smash[t]C_Gˆ(n)$ gives C(n)

G . This mayresult in a part of the top of the expression ‘smashinginto’ the characters in the line above. A slight reword-ing may prevent this.

To display or not to display

Simple expressions should not be displayed unlessyou need to give them an equation number. Morecomplicated expressions (without an equation num-ber) should be displayed if not doing so makes themdifficult to read. E.g.,

Compare $\sum_r=1ˆ\infty rˆ-z$ with\[\sum_r=1ˆ\infty rˆ-z.\]

gives Compare∑∞r=1 r

−z with

∞∑r=1

r−z.

Note that even if an expression is displayed, it is stillpart of a sentence and therefore may require punctua-tion. Punctuation marks should be placed outside of$ $ unless the marks are part of the mathematicalexpression. Also, never leave a blank line before anequation. Only leave a blank line after an equationif the equation ends a paragraph (which is unusual).E.g.,

The energy $E$ is given by\[Eˆ2 = mˆ2 cˆ4+pˆ2cˆ2,\]where $m$, $p$, and $c$ are the rest mass,momentum, and speed of light, respectively.

gives The energy E is given by

E2 = m2c4 + p2c2,

where m, p, and c are the rest mass, momentum, andspeed of light, respectively.

Never use \displaystyle within the text.This may cause uneven line spacing which looks ugly.

Functions

Functions denoted by a single letter of the (roman)alphabet, excluding any sub- or superscripts, shouldappear in italics. E.g., the associated Legendre func-tion: $P_lˆm(x)$ gives Pml (x). Functions denotedby more than one letter (such as ln or sin) shouldappear in a roman (i.e., not a slanted or italic) font(so not as ln or sin). Commonly occurring functionshave already been defined in standard LATEX or inscias.cls.

Trigonometric, hyperbolic, and exponential func-tions and their inverses \sin, \cos, \tan,\csc, \cosec, \sec, \cot, \arcsin, \arccos,\arctan, \sinh, \cosh, \tanh, \cosech,\sech, \coth, \exp, \log, \ln. [Note: log with-out a subscript always refers to the natural logarithm.If you want the base-10 logarithm you must writelog10.] With functions such as these you should followthe usual convention of bracketing arguments onlywhen necessary. E.g., $\sin 2x=2\sin x\cos x$

gives sin 2x = 2 sinx cosx and its meaning is clear.To obtain functions, such as hypergeometric func-tions, where a subscript comes first, place an emptypair of braces before the subscript. E.g., $_2F_1$gives 2F1.

www.scienceasia.org

6 ScienceAsia documentation: scias latex.pdf

Special functions \erf, \erfc, \sinc.

Functions used in analysis \sgn, \max, \min,\argmax, \argmin, \lim, \liminf, \limsup,\sup, \inf, \essinf, \supp, \hom, \deg,\Int, \ext. Note that \Int, the interior of aset, gives the function name in lower case. E.g.,$\Int(S)=S\setminus\pd S$ gives int(S) = S\∂S.

Functions used in complex analysis \Re, \Im,\arg, \Arg, \Log, \Ln, \Res.

Functions used in linear algebra \ker, \det, \tr,\diag, \adj, \dim, \rank, \spec.

Functions used in number theory \gcd, \lcm,\pmod , \bmod, \Div. \pmod and \bmodare for the modulo operation written with and withoutparentheses, respectively. E.g., $-1\bmod 3=2$ gives−1 mod 3 = 2. E.g., $x\equiv y\pmod3$ givesx ≡ y (mod 3). Note that \Div, integer division,gives the function name in lower case (as is usual).E.g., $5 \Div 3=1$ gives 5 div 3 = 1.

Functions used in statistics \Pr, \Var, \Cov.

Defining your own functions If the function youwant is not available you can define it yourself. E.g.,putting

% per A = permanent of matrix A\newcommand\per\operatornameper

somewhere before \begindocument defines afunction called per. E.g., $\per H$ gives perH . Thisis of course not necessary for functions whose symbolis a single character (such as the Bessel functionsJn) since they are always written in italics and theirarguments are always in brackets. To obtain a functionwhose sub- and superscripts appear directly belowor above the function name when used in displayedmaths, use \operatorname* instead. E.g., afterdefining

\newcommand\barlim\operatorname*\overlinelim

e.g.,

\[\barlim_x\to0f(x)=g(x).\]

giveslimx→0

f(x) = g(x).

Square roots and nth roots $\sqrtx$ gives√x,

$\sqrt[n]y$ gives n√y, $\surdz$ gives

√z.

Symbols

Defining your own roman multi-character symbolsAs with function names, parameter symbols con-taining more than one letter (excluding sub- or su-perscripts) should be in roman font. This includesparameter symbols such as SNR which are derivedfrom an abbreviation of the name of the parameter. Todefine, for example, the two letter symbol Fr, place

\newcommand\Fr\mathrmFr % Froude number

before \begindocument. Then $\Fr$ gives Fr.

Symbols involving roman sub- or superscripts If thesubscript or superscript is a letter (or letters) abbreviat-ing a word and is not itself a variable then it should bein roman font. Roman subscripts and superscripts aremost conveniently done using the commands \rs and \rp , respectively. E.g., initial temperature:$T\rsi$ gives Ti. E.g., the transpose of a matrix A:$A\rpT$ gives AT. If you need a mix of roman anditalic in the sub- or superscript then you will need touse the usual method of obtaining sub- or superscriptsand enclosing the parts that should be in roman font in\mathrm . E.g., $x_\mathrmci$ gives xci.

Greek letters

\[\alpha\beta\gamma\delta\epsilon\varepsilon\zeta\eta\theta\iota\kappa\lambda\mu\nu\xi\pi\rho\sigma\tau\phi\varphi\chi\psi\omega\Gamma\Delta\Theta\Lambda\Xi\Pi\Sigma\Upsilon\Phi\Psi\Omega\]

gives

αβγδεεζηθικλµνξπρστφϕχψωΓ∆ΘΛΞΠΣΥΦΨΩ

Calculus

\[\partial\nabla\int\oint\iint\iiint\dashint\]

gives

∂∇∫ ∮ ∫∫ ∫∫∫

−∫

Note that \dashint is the notation the journalprefers for the Cauchy principal value of an integral.

Binary operations

\[\pm\mp\times\wedge\oplus\otimes\odot\circ\ast\star\vee\]

gives±∓×∧⊕⊗ ∗ ?∨

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 7

Sets

\[\mid\emptyset\cap\cup\setminus\in\notin\subset\subseteq\supset\supseteq\forall\exists\aleph\neg\]

gives| ∅∩∪\ ∈/∈⊂⊆⊃⊇ ∀∃ℵ¬

You may wish to use \Rset, \Cset, \Zset, \Nset,and \Qset as a quick way to get R, C, Z, N, and Q,respectively.

Relations

\[\ll\leq\geq\gg\equiv\sim\simeq\gtrsim\lesssim\approx\cong\ncong\neq\propto\to\mapsto\iff\Rightarrow\Leftrightarrow\twoheadrightarrow\leftrightarrow\vdash\nvdash\downarrow\uparrow\]

gives

<<6>>>≡∼'&.≈∼= 6=∝→7→⇐⇒⇒⇔↔`0↓↑

Ellipses (lines of 3 dots)

\[\dots\ldots\cdots\vdots\ddots\iddots\]

gives

. . . . . . · · ·.... . . . .

.

You should use \dots rather than ... . E.g.,$a_0+a_1x+\dots+a_nxˆn$ gives a0 + a1x + · · · +anx

n. The dots between comma-separated objectsshould be ‘lower dots’ and those between binary oper-ations, multiplications, or integrals should be ‘centraldots’. Ideally, the \dots command should detectwhether to place lower or central dots, but it often failsto do this. You can specify lower or central dots using\ldots and \cdots, respectively.

Symbols used in QM and other branches of physics$\hbar\dagger\ell\perp\parallel$ gives ~ † ` ⊥‖.

Degrees symbols To obtain degrees Celcius use\degC after the number. E.g., 22\degC gives 22 °C.Notice that this is not in math mode. The degreessymbol on its own is obtained using \degree. E.g.,45\degree gives 45°.

Other symbols $\cdot\infty\e\ii\ij\nmid$ gives·∞ eij -. Note that we use \e to denote the base ofnatural logarithms, and \ii (or \ij, although thejournal favours i) to denote

√−1. $\male\female$

gives .

Comprehensive list See ftp://ctan.tug.org/pub/ctan/info/symbols/comprehensive/symbols-a4.pdf for acomplete list of available symbols. Note that for somesymbols you may need to include an extra package.

Making your own symbols If the symbol you wantdoesn’t exist, you might be able to create yourown by combining existing symbols. One wayis to put one symbol above another symbol us-ing \overset . E.g., $\oversetn\sim$gives n∼. You can use negative space to draw onesymbol over another such as when you need to puta diagonal line through a symbol. E.g., $A\!\!\!/$gives A/. Whenever you create a symbol you shouldmake a new command for it (see later).

Other fonts To obtain a caligraphic (blackboard)style letter, place the letter inside \mathcal (\mathbb ). E.g., $\mathcalC$ gives C. E.g.,$\mathbbB$ gives B.

Accents

Single characters or symbols may be modified us-ing the accent commands \dot , \ddot ,\dddot , \ddddot , \hat , \check ,\tilde , \bar , \breve . E.g.,

\[\dotx\ddotx+\dddotx+\ddddotx+\hatx\checkx\tildex\barx\brevex.\]

givesxx+ ˙x+ ˙ ˙x + xxxxx.

To get a line above the whole of an ex-pression (rather than a short bar of fixed length)use \overline . E.g., $\overlinexˆ2+yˆ2$

gives x2 + y2. Note the difference between,e.g., $\baru_0$ which gives u0 and, e.g.,$\overlineu_0$ which gives u0.

Space and text

To put a space between two displayed expressions onthe same line use \qquad. E.g.,

\[u_t+uˆpu_x=0, \qquad p>0.\]

givesut + upux = 0, p > 0.

A smaller (larger) gap is obtained using \quad(\qqquad or \qqqquad) instead. LATEX normallydoes a good job with formatting equations. If you needto make small adjustments, a thin space is obtainedusing \, and \; gives a slightly thicker one. A small

www.scienceasia.org

8 ScienceAsia documentation: scias latex.pdf

amount of space is deleted using \! and this is oftenused with integrals (see later).

If you need to use ordinary text in displayedexpressions put it inside \text . E.g.,\[u_t+uˆpu_x=0, \quad \textwhere $p>0$.\]

givesut + upux = 0, where p > 0.

Be aware, however, that text is often not reallyneeded, as in the above example – the meaning isclear without ‘where’. Notice that it often easier toplace text-style maths expressions inside \text as well, rather than adding another \quad after theword. Do not use \text to make functions orsymbols upright as they will be in the wrong font.Instead, define a new operator or use \mathrm .

Fractions

Only use \frac in the text if the numeratorand denominator are small positive integers. Oth-erwise use / (with brackets if necessary). E.g.,$\frac23$, $a/2(b+c)$ gives 2

3 , a/2(b + c).In displayed expressions where the numerator anddenominator are single digits, a text-style fraction ob-tained using \tfrac sometimes looks better.E.g.,\[\sin\frac12(a+b)+\cos\tfrac12(a+b).\]

gives

sin12

(a+ b) + cos 12 (a+ b).

If an expression such as a sum or integral which looksbetter in display style is in the numerator or denom-inator you should use \dsfrac instead of\frac . E.g.,\[\frac\int_-1/2ˆ1P\sodx\sum_jQ_j,\qquad\dsfrac\int_-1/2ˆ1P\sodx\sum_jQ_j.\]

gives ∫ 1

−1/2P dx∑

j Qj,

∫ 1

−1/2

P dx∑j

Qj.

Use \cfrac to get continued fractions.E.g.,\[\pi=3+\cfrac17+\cfrac115+\cfrac11+\cfrac1292+\ddots.\]

gives

π = 3 +1

7 +1

15 +1

1 +1

292 +. . .

.

Brackets and other delimiters\[()[]\\|\|\langle\rangle\lfloor\rfloor\lceil\rceil\lvert\rvert\lVert\rVert\]

gives()[]|‖〈〉bcde||‖‖

To save typing you might prefer to use \abs ,\norm , \av \floor , and \ceil ,instead of \lvert \rvert, \lVert \rVert,\langle \rangle, \lfloor \rfloor, and\lceil \rceil, respectively.

Ideally, brackets and norms in displayed expres-sions should be of about the same height as the tallestobject they enclose. If the delimiter is used on its own,the smallest size delimiter is used which gives uglyresults if the enclosed expression is much larger. E.g.,\[(\fracx1+x)ˆ2\;\textlooks ugly.\]

gives(

x

1 + x)2 looks ugly.

To automatically get the delimiterswhich are at least as high as the enclosedexpression, put \left and \right in frontof both delimiters. To save typing you mightprefer to use \lrp , \lrs , \lrb ,\lrm , \lrn , \lra , \lrf ,and \lrc instead of \left( \right),\left[ \right], \left\ \right\,\left| \right|, \left\| \right\|,\left\langle \right\rangle,\left\lfloor \right\rfloor, and\left\lceil \right\rceil, respectively.E.g.,\[\left(\fracx1+x\right)ˆ2+\lrp\fracx1+xˆ2+\lra\frac11-xˆ2+\lrf\frac11+xˆ2.\]

gives(x

1 + x

)2

+(

x

1 + x

)2

+⟨

11− x2

⟩+⌊

11 + x2

⌋.

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 9

Note that the ‘recommended’ way to obtain themodulus and norm symbols is to use, respectively,\lvert and \rvert rather than two | and \lVertand \rVert rather than two \|. This is what isdone in the definitions of \lrm and \lrn .However, in many instances the difference betweenthe two ways is not noticeable.

For every \left there must be a \right on thesame line (i.e., before the end of the environment or\\, whichever comes first). To get just one delimiteron a line, use a dot instead of the missing delimiter.E.g.,\[\left.\frac11+xˆ2\right|_0ˆ\infty.\]

gives1

1 + x2

∣∣∣∣∞0

.

Sometimes using \left and \right doesn’tgive satisfying results. To specify exactly whichsize of bracket you want, replace \left by \bigl,\Bigl, \biggl, or \Biggl and replace \rightby \bigr, \Bigr, \biggr, or \Biggr to ob-tain delimiters of increasing size. Alternatively, the\lrp , etc., commands have an optional numericargument ranging from 1–5 to specify the size of thedelimiters (1 corresponds to the smallest size). Whenonly one delimiter is required, precede the delimiterby one of the ‘big’ commands without the final ‘l’ or‘r’, e.g., \bigg. The following example shows howspecifying the delimiter size can give nicer lookingresults.\[\lrs\sum_ia_i\lrm\sum_ijx_ijˆpˆ1/p\!\!\!\!\!\!\!\!+\lrs[3]\sum_ia_i\lrm[2]\sum_ijx_ijˆpˆ1/p\!\!\!\!\!+\frac11+xˆ2\bigg|_0ˆ1.\]

gives∑i

ai

∣∣∣∣∣∣∑ij

xij

∣∣∣∣∣∣p1/p

+[∑i

ai∣∣∑ij

xij∣∣p]1/p+ 1

1 + x2

∣∣∣∣10

.

The \mid symbol sometimes used for ‘such that’cannot be resized. Use \;\bigg|\; or \biggmidinstead. E.g.,\[\lrb\beginpmatrixe&f\\kf&e\endpmatrix\biggmid e,f\in\Nset\]

gives (e fkf e

) ∣∣∣∣ e, f ∈ N

Derivatives and integrals

The (ordinary) differential ‘d’ should be in a romanfont. The easiest way to ensure this is to use thecommand \od to obtain the differential. E.g.,$\ody/\odx$ gives dy/dx. For displayed firstderivatives and nth derivatives you may find it easierto use \fod and \fodn , respec-tively. E.g.,

\[\fodnyx2+\fodyx\]

givesd2y

dx2+

dydx

although the ‘dash notation’ (i.e., $y’’+y’$ gives y′′+y′) in such a case is often preferable since it takes upless space. Similarly, for time derivatives use the ‘dotnotation’ (e.g., $\ddoty+\doty$ gives y+ y).

For partial derivatives you may prefer to use \pdinstead of \partial to save typing. Displayedpartial derivatives are easily done using \fpd and \fpdn . E.g.,

\[\fpdnt=\fpdnnx2.\]

gives∂n

∂t=∂2n

∂x2.

However, if you are using partial derivatives a lotthen you should use the suffix notation instead (e.g.,$n_t=n_xx$ gives nt = nxx).

For integrals there should be a space between theintegrand and the differential. This can be done byplacing the integration variable inside \sod . E.g.,

\[\int\sin x\sodx.\]

gives ∫sinxdx.

If the integral has limits you may find you need touse a few \! before the integrand for the best results.E.g.,

\[\int_-\inftyˆ\infty\!\!\!\!\sechˆ2x\sodx.\]

gives ∫ ∞−∞

sech2 xdx.

www.scienceasia.org

10 ScienceAsia documentation: scias latex.pdf

Sums, unions, and products

To form a sum, product, or union use the commands\sum, \product, and \bigcup and specify thelimits using sub- and superscripts. E.g.,

\[\det M=\prod_i=1ˆn\lambda_i.\]

gives

detM =n∏i=1

λi.

To obtain multiline limits use \substack insidethe sub- or superscript and specify new lines using \\.E.g.,

\[\Phi_j=-\sum_\substacki=1\\i\neq jˆn\fracGm_ir_ij.\]

gives

Φj = −n∑i=1i6=j

Gmi

rij.

The following symbols can be used in a similarway.

\[\coprod\bigcap\bigvee\bigwedge\bigoplus\bigotimes\bigodot\]

gives ∐⋂∨∧⊕⊗⊙Matrices and other arrays of quantities

Place elements of a matrix between\beginpmatrix (or \bpm) and\endpmatrix (or \epm). Separate elements onthe same row by & and put a \\ at end of each rowexcept the last row. E.g.,

\[\beginpmatrixa&b&c\\d&e&f\\g&h&i\endpmatrix.\]

gives a b cd e fg h i

.

Use \beginvmatrix (or \bvm) and\endvmatrix (or \evm) to obtain determinantsin a similar way. E.g.,

\[\beginvmatrixA_11 & A_12 & \cdots & A_1n \\A_21 & A_22 & \cdots & A_2n \\\vdots & \vdots & \ddots & \vdots \\A_n1 & A_n2 & \cdots & A_nn\endvmatrix.\]

gives ∣∣∣∣∣∣∣∣∣A11 A12 · · · A1n

A21 A22 · · · A2n

......

. . ....

An1 An2 · · · Ann

∣∣∣∣∣∣∣∣∣ .The elements of matrices and determinants are

centred when done this way. This might not look sogood if some of the elements have minus signs. Thereare two possible ways to deal with this. The first is toadd invisible characters to balance the element usingthe \phantom command. E.g.,

\[\bpm -1&2\\ 3&4\epm\bpm -1&2\\ \phantom-3&4\epm.\]

gives (−1 23 4

)(−1 2

3 4

).

The other way is to replace the begin andend matrix commands by \beginarray and \endarray and enclose these commandswith the appropriate delimiters. The argument of\beginarray must contain n characters ifthe matrix has n columns. Each character can be l,c, or r which correspond to left, centre, and rightjustification, respectively. E.g.,

\[\lrp\beginarrayrc-1&2\\3&4\endarray.\]

gives (−1 2

3 4

).

For binomial coefficients, rather than constructinga 2×1 matrix, you can just use \tbinom forbinomial coefficients in the text and \binom for displayed binomial coefficients. E.g.,$\tbinom42$ gives

(42

). E.g.,

\[\binomnk.\]

gives (n

k

).

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 11

An array of quantities with no delimitersis obtained using \beginmatrix and\endmatrix. An array of quantities enclosed inbraces is obtained using \beginBmatrix and\endBmatrix. These are useful for giving arraysof alternative functions. E.g.,\[\beginmatrixa_m\\b_m\endmatrix\biggr\=\frac1\pi\int_0ˆ2\pi\!\!\!\! f(x)\beginBmatrix\cos mx\\\sin mx\endBmatrix\sodx.\]

gives

ambm

=

1π

∫ 2π

0

f(x)

cosmxsinmx

dx.

Piecewise functions

For functions whose expression depends on the ar-gument, treat the expressions and conditions likeelements in a matrix but instead of the begin andend matrix commands use \begincases and\endcases. E.g.,\[\sgn(x)=\begincases1, & x>0,\\0, & x=0,\\-1, & x<0.\endcases\]

gives

sgn(x) =

1, x > 0,0, x = 0,−1, x < 0.

If any of the expressions require more than one linethen use \\ followed by \quad as in the followingexample.\[q(x,y)=\begincasesa+b+c+d+e\\\quad\mbox+f+g+h, & x>1,\\0, & \text$x<1$, $y=0$,\\& \textand $q=a+b+c$,\\-1, & \textotherwise.\endcases\]

gives

q(x, y) =

a+ b+ c+ d+ e

+ f + g+ h, x > 1,0, x < 1, y = 0,

and q = a+ b+ c,

−1, otherwise.

The \mbox is used to get the correct spacing afterthe + sign it precedes.

If the cases contain fractions, you normallyget better results if you replace \begincasesand \endcases by \begindcases and\enddcases. E.g.,

\[x=\begindcases\fracac(\eˆ\alpha t-1)a\eˆ\alpha t-bc, & a>bc,\\\fracac(\eˆ\alpha t-1)bc\eˆ\alpha t-a, & a<bc,\\\fracc\beta t1+\beta t, & a=bc.\enddcases\]

gives

x =

ac(eαt − 1)a eαt − bc

, a > bc,

ac(eαt − 1)bc eαt − a

, a < bc,

cβt

1 + βt, a = bc.

Vectors

Vectors (in the physical sciences, at least) should beshown using bold font (and not by underlining orusing ~ ). This is easily done using the \vect command. A unit vector (denoted by a bold fontcharacter with a hat) is obtained using \uvect .E.g.,

\[\vecta\times\vectb=\uvectnab\sin\theta.\]

givesa× b = nab sin θ.

Care is needed over what is placed in the argumentof these commands. For example, the \dot used to denote the time derivative should be ap-plied after \vect . Subscripts and superscriptsshould not appear inside these commands. E.g.,$\dot\vect\omega_i$ gives ωi.

The scalar product dot is obtained using \spdot.E.g.,

\[\vecta\spdot\vectb=ab\cos\theta.\]

givesa · b = ab cos θ.

The commands \grad, \div, and \curl havebeen defined in the expected way. E.g.,

\beginmultline*\curl(\vecta\times\vectb)\equiv(\div\vectb)\vecta-(\div\vecta)\vectb\\+(\vectb\spdot\grad)\vecta-(\vecta\spdot\grad)\vectb.\endmultline*

www.scienceasia.org

12 ScienceAsia documentation: scias latex.pdf

gives

∇× (a× b) ≡ (∇ · b)a− (∇ · a)b+ (b · ∇)a− (a · ∇)b.

For the Laplacian operator use \Delta ifyou don’t wish to use ‘del-squared’. E.g.,$\Delta\equiv\nablaˆ2$ gives ∆ ≡ ∇2.

Equations

A displayed equation should only have an equationnumber if either you refer to the equation later in themanuscript or the equation is a key result which othersmay wish to refer to when citing your article.

Numbering, labelling, and referring to equationsFor a single numbered equation which needsonly one line use \beginequation and\endequation instead of \[ and \]. If yourefer to the equation it will need a label which you canassign using \label . We strongly recommendthat the label you use for equations starts with e:and that the label you choose is meaningful (to you,at least). The worst choice for a label is an equationnumber since the actual number of the equation maychange if you later add or delete equations beforeit or if you copy and paste the equation to anotherdocument. E.g.,

\beginequation\labele:s2id\sechˆ2x=1-\tanhˆ2x.\endequation

givessech2 x = 1− tanh2 x. (1)

To refer to the equation later use \eqref . The ar-gument of this command is the same as the label usedin \label . Brackets are generated automatically.E.g., using \eqrefe:s2id gives using (1). Onlyuse the word ‘equation’ or ‘Eq.’ before \eqref if it starts a sentence.

Single multi-line equation For a single numberedequation that needs more than one line use\beginmultline and \endmultlineinstead of \beginequation and\endequation and show where the newlines are using \\. E.g.,

\beginmultline\labele:Gammanz\Gamma(z)\Gamma\lrpz+\frac1n\Gamma\lrpz+\frac2n\cdots\Gamma\lrpz+\fracn-1n \\=(2\pi)ˆ(n-1)/2nˆ\frac12-nz\Gamma(nz),\quad n=1,2,\ldots\,.\endmultline

gives

Γ(z)Γ(z +

1n

)Γ(z +

2n

)· · ·Γ

(z +

n− 1n

)= (2π)(n−1)/2n

12−nzΓ(nz), n = 1, 2, . . . . (2)

For an equation with no number that needs morethan one line use \beginmultline* and\endmultline* instead. E.g.,

\beginmultline*a+b=c-d+e+f-g+h-i\\+j-k+l-m+n\\\mbox-o+p-q+r-s.\endmultline*

gives

a+ b = c− d+ e+ f − g+ h− i+ j − k+ l−m+ n

− o+ p− q+ r− s.

Note that with equations that need more than one line,no line should end with =, +, or −. Such symbolsshould be moved to the next line. Also, in general, ifa line starts with a − then it should be preceded by\mbox in order to get the correct spacing after the− sign, as in the example above.

Saving space In cases such as (2) where the expres-sion(s) almost fill the line, there are two things you cando to help. First, just before \beginmultline or\begingathered place a \zmlg (which standsfor zero multline gap). This will cause the first partof the equation to be left justified. Second, use \! toremove unnecessary space. E.g.,

\zmlg\beginmultline\labele:Gammanz_\Gamma(z)\,\Gamma\!\lrp\!z+\frac1n\Gamma\!\lrp\!z+\frac2n\cdots\Gamma\!\lrp\!z+\fracn-1n \\=(2\pi)ˆ(n-1)/2nˆ\frac12-nz\Gamma(nz),\quad n=1,2,\ldots\,.\endmultline\rmlg

gives

Γ(z) Γ(z +

1n

)Γ(z +

2n

)· · ·Γ

(z +

n− 1n

)= (2π)(n−1)/2n

12−nzΓ(nz), n = 1, 2, . . . . (3)

The multline gap can be restored to its usual valueby adding a \rmlg (= restore multline gap) after\endmultline.

Set of aligned equations If you have a set of similarequations of similar size or content it looks betterif the equals signs (or other relational operators) are

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 13

aligned. If they are numbered equations, start andend the set of equations with \beginalign and\endalign and show the alignment by putting an& before the character you wish to align. E.g.,

\beginalign\labele:2sc2\sin x\cos y&=\sin(x+y)+\sin(x-y),\\\labele:2cc2\cos x\cos y&=\cos(x+y)+\cos(x-y),\\\labele:2ss2\sin x\sin y&=\cos(x-y)-\cos(x+y).\endalign

gives

2 sinx cos y = sin(x+ y) + sin(x− y), (4)2 cosx cos y = cos(x+ y) + cos(x− y), (5)2 sinx sin y = cos(x− y)− cos(x+ y). (6)

If the equations are not to be numbered, use\beginalign* and \endalign* instead.E.g.,

\beginalign*\sin(x\pm y)&=\sin x\cos y\pm\cos x\sin y,\\\cos(x\pm y)&=\cos x\cos y\mp\sin x\sin y.\endalign*

gives

sin(x± y) = sinx cos y± cosx sin y,cos(x± y) = cosx cos y∓ sinx sin y.

If an equation which needs to be aligned needsmore than one line, break it in the usual way using\\ and follow this by & and then \qquad. If theequation is numbered, you need to put \notag (or\nonumber) before the \\ to prevent a numberappearing at that line. Take care to place the labelcommand in a part which does have a number. E.g.,

\beginalignP(x)&=a+b+c+d+e+f+g\notag\\\labele:p&\qquad+h+i+j+k+l,\\\labele:qq&=r+s+t.\endalign

gives

P (x) = a+ b+ c+ d+ e+ f + g

+ h+ i+ j + k+ l, (7)q = r+ s+ t. (8)

Set of unaligned equations If the equations donot need to be aligned, use \begingatherand \endgather instead of \beginalignand \endalign (or \begingather* and\endgather* if you don’t need equation num-bers) and omit the & everywhere. E.g.,

\begingather\labele:b+cb+c=p+q+r,\\\labele:QQ=d+e+g+h+i+j+k+l+m+n.\endgather

gives

b+ c = p+ q+ r, (9)Q = d+ e+ g+ h+ i+ j + k+ l+m+ n. (10)

One equation number for several equations on sep-arate lines The equations are separated by \\ andtheir alignment can be specified using & as usual. Theequations are placed within \beginsplit and\endsplit. The split environment is placed in anequation environment where the equations are giventheir label. E.g.,\beginequation\labele:sim\beginsplitx+y&=6+p,\\2x+y+z&=7.\endsplit\endequation

gives

x+ y = 6 + p,

2x+ y+ z = 7.(11)

If the & are omitted the equations are right jus-tified. If no alignment or justification is wanted,use \begingathered and \endgatheredinstead of the split environment. E.g.,\beginequation\labele:lap\begingathered\nablaˆ2u(\vectr)=0,\quad\vectr\in S,\\u(\vectr)=0,\quad\vectr\in\pd S.\endgathered\endequation

gives∇2u(r) = 0, r ∈ S,u(r) = 0, r ∈ ∂S.

(12)

If you wish to give a single equation number to morethan two lines of equations (and in particular if thereare an odd number of lines) then you may wish to adda brace on the right-hand side to make this clearer.If the equations are aligned then place them between\beginaligned and \endaligned as in thefollowing example. E.g.,\beginequation\labele:ACE\left.\beginalignedA&=B+a+b+c+d+e+f,\\C&=(D+g+h+i+j+k)\\&\qquad\times(r+s+t+u+v+w),\\E&=F+l+m+n+o+p+q.\endaligned\right\\endequation

www.scienceasia.org

14 ScienceAsia documentation: scias latex.pdf

gives

A = B + a+ b+ c+ d+ e+ f,

C = (D+ g+ h+ i+ j + k)× (r+ s+ t+ u+ v+w),

E = F + l+m+ n+ o+ p+ q.

(13)

If the equations do not need to be aligned thenuse \begingathered and \endgathered.E.g.,\beginequation\labele:PQR\left.\begingatheredP=B+a+b+c+d+e+f,\\Q=D+g+h+i+j+k,\\R=p+q.\endgathered\right\\endequation

gives

P = B + a+ b+ c+ d+ e+ f,

Q = D+ g+ h+ i+ j + k,

R = p+ q.

(14)

Multi-line partially aligned derivation The com-mand structure given here needs to be used if the firstline should not be aligned with the equals signs in thelines below. E.g.,\beginalign*&a+b+c+d+e+f+g+h\\&\quad=p+q+r+\biggl(\fracs2+t+l+m+n\\&\qqquad+P+Q+R\biggr)ˆ1/2+A+B+C\\&\quad=u+v+w.\endalign*

gives

a+ b+ c+ d+ e+ f + g+ h

= p+ q+ r+(s

2+ t+ l+m+ n

+ P +Q+R

)1/2

+A+B +C

= u+ v+w.

If an equation number is required then do the follow-ing. E.g.,\beginalign&a+b+c+d+e+f+g+h\notag\\&\quad=p+q+r+s+t+l+m+n\notag\\&\qqquad+P+Q+R+A+B+C\notag\\&\quad=u+v+w.\labele:uvw\endalign

gives

a+ b+ c+ d+ e+ f + g+ h

= p+ q+ r+ s+ t+ l+m+ n

+ P +Q+R+A+B +C

= u+ v+w. (15)

Subequations The idea of subequations is togive a whole set of equations a separate labelfrom the individual equations in the set. Thisis done by placing the complete set of labelledequations (which could be in an align or gatherenvironment) between \beginsubequationsand \endsubequations. The label forthe whole set of equations is placed after\beginsubequations but before the beginningof the environment giving the set of equations. E.g.,\beginsubequations\labele:lorenz\beginalign\labele:lorenzx\dotx&=\sigma(y-x), \\\labele:lorenzy\doty&=rx-y-xz, \\\labele:lorenzz\dot z&=xy-bz.\endalign\endsubequations

gives

x = σ(y− x), (16a)y = rx− y− xz, (16b)z = xy− bz. (16c)

Only use subequations if you refer to both the set ofequations as a whole, e.g. (16), and at least one of thesubequations individually, e.g. (16c).

Allowing sets of equations to be split across pagesor columns If you wish to allow a set of equa-tions or a multiline equation to be split acrossa page or column you will need to enclose itwith \allowdisplaybreaks . Note that thisshould be used for each equation which is causingproblems; \allowdisplaybreaks should not beplaced in the preamble. The closing brace should beplaced after text rather than at end of the displayedequation. This closing brace must be immediately fol-lowed by a % to avoid a rogue space on the followingline. Note that even if this command is used, equationsinside the split, aligned, and gathered environmentscannot be split across a page or column.

CHEMISTRY EXPRESSIONS

The command \ce is a convenient way to obtainchemical formulae (including those specifying thetype of bond) and chemical (or nuclear) reactions.For more details on this command than are givenhere see ftp://www.ctan.org/tex-archive/macros/latex/contrib/mhchem/mhchem.pdf.

Formulae

Compounds No underscores are neededif the subscripts are single digits. E.g.,

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 15

\ceH2SO4 gives H2SO4. E.g., \ce(CH3)2S

gives (CH3)2S. In other cases subscriptsmust be specified in the usual way. E.g.,\ceAl_0.25Ga_0.75As gives Al0.25Ga0.75As.E.g., \ceAl_$x$Ga_$1\!-\!x$As givesAlxGa1−xAs. Water of crystallization is added after adot. E.g., \ceCuSO4.5H2O gives CuSO4 · 5 H2O. Nofont change commands are needed for \ce insidemath mode. E.g., $\epsilon_\ceCO2$ gives εCO2

.

Ions To get a singly charged ion just add + or −to the end. E.g., \ceNO3- gives NO–

3. If the ion ismore than singly charged you need a ˆ before the sizeof the charge. E.g., \ceSO4ˆ2- gives SO2–

4 . E.g.,\ce[Zn(OH)4]ˆ2- gives [Zn(OH)4]2–.

Radicals For neutral radicals add a ˆ. (e.g.,\ceOHˆ. gives OH·). For charged radicals thecharge and dot need to be enclosed in braces. E.g.,\ceRNO2ˆ-. gives RNO–·

2 .

Isotopes For isotopes place the atomic and massnumbers as sub- and superscripts before the element.E.g., \ce_92ˆ235U gives 235

92U.

Bonds Single, double, and triple bonds are denotedby −, =, and #, respectively, if placed between twoelements. E.g., \ceA-B=C#D gives A−B−−C−−−D.These need to be preceded by if placed at the startof the expression. E.g., \ce-NH2 gives −NH2.

Reactions

A whole reaction may be placed inside \ce . Thereaction may be placed in the text or in displayed forminside \[ and \] or inside an equation environment.The possible types of arrow are shown in the followingexample.

\[\ce-> <- <-> <=> <=>> <<=> ˆ v\]

gives

−−→←−− ←−→ −−−− −−− −−− ↑ ↓

The horizontal arrows have two optional arguments,placed in square brackets immediately after the arrowsymbol. The first is the expression above the arrow;the second is for below. Both expect math modeexpressions so you will need to use \text if youwish to place words there. E.g.,

\[\ceH+ + OH- <=>>[k_1][k_-1] H2O\]

gives

H+ + OH− −−−k−1

k1−−−−− H2O

Note that the addition +must be surrounded by spaces.If you wish to show a series of reactions and

align them using & and \\ in an aligned equationenvironment, replace \ce by \cee .

Symbols

The standard state symbol is obtained using \stst.E.g., $\Delta H\stst$ gives ∆H− .

FLOAT ENVIRONMENTS

All float environments (i.e., tables, figures, and list-ings) should contain a caption and a label. In thecase of tables and figures, the caption is placed in\caption and the label is placed in \label so that it can be referred to using \ref . Bydefault, floats will be positioned at the top of thepage (or immediately below another float). Theycan be forced to appear at the bottom of a page byputting [b] immediately after the begin environmentcommand.

Tables

Floating tables are placed inside \begintableand \endtable. After the caption and label,the tabulated part is placed between \btsf and\etsf. The argument of \btsf is a list ofcharacters, one for each column, giving the type ofjustification for the corresponding column: l = left;c = centre; r = right. The first column normallylooks best with left justification. Elements in thetabular array are separated by & and a \\ marks theend of each row. The headings are separated fromthe entries below by a horizontal line made using\midrule. A slightly thinner line across columnsc1 to c2 is obtained using \cmidrulec1-c2. Itcan be shortened on the left, right, or both by adding(l), (r), or (lr), respectively, before the argumentopening brace. Columns on the same row can bemerged using \multicolumn . The firstargument is the number of columns to be merged,the second is the justification character, and the thirdis the contents of the merged entries. Note that\multicolumn1 can be used to changethe justification of a single entry. To add an extra gapbetween rows use \addlinespace. Avoid usingother formatting techniques such as adding emptycolumns or explcitly giving column widths.

If there are table notes these are placed between\btn and \etn. If they refer to specific symbolsin the tabular region (placed as superscripts using

www.scienceasia.org

16 ScienceAsia documentation: scias latex.pdf

\tnote ) then the symbol should be placed insquare brackets after an \item command. E.g.,\begintable % 1-column table\captionLabelling conventionfor environments.\labelt:envlab\btsflclcenviron- & label &\multicolumn2cstart of reference \\\cmidrule(lr)3-4ment& start & \multicolumn1clong form& short form \\ \midruleequation\tnotea &\Te&\tce & \Tce \\table & \Tt & \tcTablet & \Tct \\figure & \Tf & \tcFig.f & \Tcf \\lstlisting &\Tl&\tcListingl & \Tcl \\\addlinespaceitem\tnoteb &\Ti&\texttt(\textbackslashref\i: & \Tcei \\item\tnotec &\Ti&\texttt\textbackslashref\i: & \Tci \\\addlinespacetheorem & \TT & \tcTheoremT & \TcT \\lemma & \TL & \tcLemmaL & \TcL \\corollary &\TC&\tcCorollaryC & \TcC\\proposition&\T P&\tcPropositionP&\TcP\\identity &\TI&\tcIdentityI& \TcI\\conjecture &\TJ&\tcConjectureJ& \TcJ\\remark & \TR&\tcRemarkR& \TcR\\definition &\TD&\tcDefinitionD&\TcD\\example & \TE & \tcExampleE & \TcE \\algorithm &\TA&\tcAlgorithmA & \TcA\\\etsf\btn\item[a] and other environments withequation numbers\item[b] inside enumerate environment\item[c] inside steps environment\etn\endtable

gives [see Table 1]. To refer to a table in the textuse \ref . E.g., Table˜\reft:envlab gives Ta-ble 1. Alternatively, if you use \tref you don’tneed to type Table or t: and the hyperlink to the tablewill include the word Table. E.g., \trefenvlab

gives Table 1.A table spanning both columns is obtained by

using \begintable* and \endtable* in-stead.

Figures

The information for a figure is placed between\beginfigure and \endfigure. As thecaption appears below, \caption followed by\label are placed after the graphics. Thesimplest way to input a graphics file is to use\incgcw whose argument is the filename. Thismakes the width of the figure match the columnwidth. If this makes the figure appear too large, use\cincgcm instead. Its first argument is thewidth in cm, and the second is the filename. In thecase of PDF files (or encapsulated PostScript files ifyou are creating a .ps file) the final .pdf (or .eps) of thefilename may be omitted. The code used to produceFig. 1 is shown below.

Table 1 Labelling convention for environments.

environ- label start of reference

ment start long form short form

equationa e: \eqrefe: \ereftable t: Table˜\reft: \treffigure f: Fig.˜\reff: \freflstlisting l: Listing˜\refl: \lref

itemb i: (\refi: \eirefitemc i: \refi: \iref

theorem T: Theorem˜\refT: \Treflemma L: Lemma˜\refL: \Lrefcorollary C: Corollary˜\refC: \Crefproposition P: Proposition˜\refP: \Prefidentity I: Identity˜\refI: \Irefconjecture J: Conjecture˜\refJ: \Jrefremark R: Remark˜\refR: \Rrefdefinition D: Definition˜\refD: \Drefexample E: Example˜\refE: \Erefalgorithm A: Algorithm˜\refA: \Aref

a and other environments with equation numbersb inside enumerate environmentc inside steps environment

\beginfigure\incgcwscias_template\captionOutput PDF file from running latexon a copy of \filenamescias\_template.tex.\labelf:scias_template\endfigure

To refer to a figure in the text use \ref or\fref . E.g., Fig.˜\reff:scias_template

gives Fig. 1. With \fref the hyperlink includesthe word Fig. E.g., \frefscias_template givesFig. 1.

A figure spanning both columns is obtained byusing \beginfigure* and \endfigure*instead.

Program listings

Code is placed between \beginlstlistingand \endlstlisting. In this case the captionand label are done differently – see the example below.E.g.,\beginlstlisting[float=t,caption=C function computing the rootsof $axˆ2+bx+c=0$, where $a,b,c\in\bbR$.,label=l:quadroots]void quadroots(double a,double b,double c,double *re1,double *im1,double *re2,double *im2) double q,dis=b*b-4*a*c;if (dis>0) q=(b>0?-sqrt(dis)-b:sqrt(dis)-b)/2;*re1=q/a;*re2=c/q;*im1=*im2=0;

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 17

Listing 1 C function computing the roots of ax2+bx+c =

0, where a, b, c ∈ R.

1 void quadroots2 (double a,double b,double c,3 double *re1,double *im1,4 double *re2,double *im2) 5 double q,dis=b*b-4*a*c;6 if (dis>0) 7 q=(b>0?-sqrt(dis)-b:sqrt(dis)-b)/2;8 *re1=q/a;9 *re2=c/q;

10 *im1=*im2=0;11 else 12 *re1=*re2=-b/a/2;13 *im2=-(*im1=sqrt(-dis)/a/2);14 15

else *re1=*re2=-b/a/2;*im2=-(*im1=sqrt(-dis)/a/2);

\endlstlisting

gives [see Listing 1]. If you wish the listing to appearat the bottom of the page replace float=t by float=b.The two-column version is obtained by placing a *before the t or b.

To refer to a listing in the text use \ref or\lref . E.g., Listing˜\refl:quadroots givesListing 1. With \lref the hyperlink includesthe word Listing. E.g., \lrefquadroots givesListing 1. If you wish to refer to a variable in the code,then enclose it in \lstinline! ! where the ! canbe replaced by any character which is not in the code.E.g., \lstinline!dis! is the discriminant givesdis is the discriminant.

Note that this environment is for showing frag-ments of code or pseudocode. It is not intended fordisplaying entire programs.

THEOREMS AND PROOFS

Theorems and similar environments

A theorem should be placed between\begintheorem and \endtheorem.The text inside the theorem environment will appearin italics. You should not change the font forthe whole theorem yourself. The label is placedafter \begintheorem using \labelT: .A theorem is referred to using \refT: or\Tref . The other theorem-like environments(Table 1) are used in an analogous way.

A theorem can be named by adding the name insquare brackets directly after \begintheorem.In environments where text is italicized automatically,you should put digits inside $ $ or \upshape tostop them appearing in italics. E.g.,

\beginlemma[Hippasus]\labelL:sqrt2The square root of $2$ is irrational.\endlemma

gives

Lemma 1 (Hippasus) The square root of 2 is irra-tional.

Sub-results in theorems starting with (i), (ii), etc.,can be given by placing them after \item. Theitems are all between \beginenumerate and\endenumerate and this environment can benested. The \item command will automaticallygenerate the number. Each item can be given a label.We suggest following the convention of starting thelabel with i:. E.g.,

\begincorollary\labelC:r2Suppose $n$ is a non-zero integer.\beginenumerate\item \labeli:nr2 $n\sqrt2$ isirrational.\item \labeli:r2/n $\sqrt2/n$is irrational. Also,\beginenumerate\item \labeli:r2/2n $\sqrt2/2n$is irrational;\item $\sqrt2/nˆ2$ is irrational.\endenumerate\item \labeli:(r2/n)ˆ2n\[\lrp\frac\sqrt2nˆ2n \quad\textis rational.\]\endenumerate\endcorollary

gives

Corollary 1 Suppose n is a non-zero integer.(i) n

√2 is irrational.

(ii)√

2/n is irrational. Also,(a)√

2/2n is irrational;(b)√

2/n2 is irrational.(iii) (√

2n

)2n

is rational.

The items may then be referred to using \ref in the usual way or \eiref . The advantageof \eiref is that it includes the bracketsand it ensures that the brackets and romannumerals inside are always upright. E.g.,from \Crefr2\eirefnr2 and \eirefr2/2n

gives from Corollary 1(i) and (ii-a). To obtain labelsin the form of capital letter and a subscripted number,

www.scienceasia.org

18 ScienceAsia documentation: scias latex.pdf

place the capital letter in square brackets immediatelyafter the \beginenumerate. E.g.,

We assume that\beginenumerate[A]\item\ilabelevenf $f(x)$ is even;\item $g(x)$ is odd.\endenumerate

gives We assume that(A1) f(x) is even;(A2) g(x) is odd.The numbering in these labels can be continued laterby using \beginenumerate[resume*=?]where ? stands for the capital letter. E.g.,

Given \eirefevenf we may also assume that\beginenumerate[resume*=A]\item $h(x)$ is odd.\endenumerate

gives Given (A1) we may also assume that(A3) h(x) is odd.

Proofs

Except when the proof is the entire section or subsec-tion (as indicated by the heading), a proof is placed be-tween \beginproof and \endproof. E.g.,

\beginproofThe proof is obvious.\endproof

gives Proof : The proof is obvious. 2If the proof does not immediately follow the

theorem, etc. then you can specify what the proof isof using the optional argument. E.g.,

\beginproof[of \Lrefsqrt2]The proof is obvious.\endproof

gives Proof of Lemma 1: The proof is obvious. 2

Algorithms

Steps of the algorithm can be given by placingeach step after an \item command all in between\beginsteps and \endsteps. The steps (oritems) may be labelled and referred to in the usualway. E.g.,

\beginalgorithm\beginsteps\item Initialize.\item \labeli:iter Iterate.\item Check accuracy. If not accurateenough go to Step˜\refi:iter.\item Save results.\endsteps\endalgorithm

gives

Algorithm 1Step 1: Initialize.

Step 2: Iterate.Step 3: Check accuracy. If not accurate enough go to

Step 2.Step 4: Save results.

The steps environment can be used elsewhere – itdoes not need to be within the algorithm environment.

FONTS AND TEXT SYMBOLS

Text may be emphasized (for example, when givinga term which is being defined) by enclosing it in\emph or \em . Text may be italicized by en-closing it in \textit or \it and made boldby enclosing it in \textbf or \bf . Smallcaps is obtained using \textsc or \sc . InScienceAsia, we show names of software in smallcaps, but this should be done by placing the name in\prog .

Single characters that have another meaning inLATEX can be obtained in the text (i.e., not as part ofa mathematical expression) by preceding them with a\. E.g., \% \# \& \_ \$ gives % # & $. If a text-symbol command ends in a letter then you need toadd a \ immediately after the command if you want aspace after the symbol. E.g., 20\degC\ for 1˜h gives20 °C for 1 h.

SHORT CUTS

These are ways to save yourself some typing and time.

Short versions of commands

Shorter alternative versions of various commandshave been defined in scias.cls (see Tables 1 and2).

Defining your own commands

Defining your own commands has two advantages.First, it will save you some typing. Second, andmore importantly, it will allow you to change notationeasily. To define a command with no arguments use\newcommand . The first argument containsthe new command and the second contains its defini-tion. An example has already been given earlier in thesection on defining your own multi-character symbols.

To define a new command with n arguments use\newcommand [n] . Again, the first argu-ment is the name and the last argument is the defini-tion. In the definition the nth argument is representedby the symbol #n. For example, suppose you find thatyou are using a lot of 3-d column vectors. You mightdecide to define a command \myv whose3 arguments are the coordinates. You could do thiswith

www.scienceasia.org

ScienceAsia documentation: scias latex.pdf 19

Table 2 Shortened versions of various commands.original form short form

\partial \pd\phantom- \phM\beginpmatrix† \bpm\beginvmatrix \bvm\beginequation \beq\beginmultline‡ \bml\beginalign \bal\beginaligned \bald\begingather \bga\begingathered \bgad\beginsplit \bsp\beginsubequations \bse\beginfigure \bfig\begintable \btab\beginenumerate \ben\labele:§ \elabel\begintheorem§§ \beginthm\beginlemma \beginlem\begincorollary \begincor\beginproposition \beginprop\beginidentity \beginidn\beginremark \beginrem\begindefinition \begindefn\beginalgorithm \beginalg

† the short form for the end environment command isobtained by replacing b by e. E.g., the short form of\endpmatrix is \epm

‡ for the starred (*) form of a command add s tothe end of the short form. E.g., the short form of\beginmultline* is \bmls

§ other short forms of \label follow the same pat-tern as for short forms of \ref – see Table 1.

§§and the short form of \endtheorem is \endthm,etc.

\newcommand\myv[3]\bpm#1\\#2\\#3\epm

and then, for example,

\[\myv45qˆ2/3.\]

gives 45q2/3

.

If you later decide to use the more compact notation ofa horizontal row of comma-separated quantities, youjust need to change the definition to (#1,#2,#3).

Note that the name of your new command canonly contain letters of the alphabet (i.e., a–z, A–Z, andno digits or other symbols).

Arguments of commands

If the argument of a command is only a single char-acter (or a single LATEX command such as \infty)it does not need to be enclosed in braces. E.g.,$\frac12$ gives 1

2 . But if the first argument is asingle letter then it must be preceded by a space. E.g.,$\sqrt n$ gives

√n.

COMMANDS NOT TO USE

Do not use any of the following commands anywherein your manuscript.

• \def \renewcommand

• \DeclareMathOperator

• \setlength

• \tag

• \dfrac \splitfrac \over

• \aligned \unaligned

• \beginsmallmatrix\endsmallmatrix

• \subfigure

• \beginsidewaystable\endsidewaystable

• \beginitemize \enditemize

• \begindescription\enddescription

INDICATING CHANGES IN THE REVISEDMANUSCRIPT

For the benefit of the referees, you might like toshow the parts of the manuscript text which havechanged since the previous version by enclosing themin \* . These parts will then appear in purple. E.g.,\*We thank the referees for their comments.

gives We thank the referees for their comments. Oncethe manuscript reaches the editing stage, thesecommands will be removed automatically (whilepreserving their arguments) by our manuscript-processing software.

SUBMISSION TO OTHER JOURNALS

If you wish to submit your manuscript preparedusing scias.cls to another journal, then using thenoscias option (i.e., starting the document with\documentclass[noscias]scias) produces an out-put file with no mention of ScienceAsia. To avoidhaving linenumbers, include nolineno as an option.

www.scienceasia.org

20 ScienceAsia documentation: scias latex.pdf

OTHER CLASS FILES

If you wish to use the non-standard or redefinedcommands explained here with other LATEX class filesyou are welcome to do so. Their definitions are in thefile www.scienceasia.org/scias macros.tex (which youare free to download) and may be included by copyingand pasting into the preamble of your document or byputting

\inputscias_macros

in the preamble after you have included the requiredpackages by placing some or all of

\usepackageamsmath,amssymb,mathdots,mathtools\usepackageenumitem\usepackagegraphicx\usepackagethreeparttable,booktabs\usepackage[version=3]mhchem\usepackageurl\usepackagelistings

in the preamble (if the packages are not alreadyincluded by the class file you are using).

Acknowledgements: We thank Paweena Kongkaew,Mongkon Youngtanurat, and Rachada Sirawaraporn fortheir invaluable help with ensuring that all aspects of thejournal run smoothly.

REFERENCES1. Lamport L (1994) LATEX: A Document Preparation Sys-

tem, 2nd edn, Addison-Wesley, Reading, MA.2. Brihaye Y, Hartmann B (2004) Fullerenic solitons. J

Phys A 37, 1181–1192.3. Sloan IH (1990) Superconvergence. In: Golberg MA

(ed) Numerical Solutions of Integral Equations, PlenumPress, New York, pp 35–70.

www.scienceasia.org