Blog

Equation Formatting

A Complete Guide for Academic and Scientific Documents

60 min read Scientific Writing & Typesetting LaTeX · Word · MathML 10,000+ words
LaTeX Word / MathType Math Notation Discipline Standards
Custom University Papers Academic Team
Expert guidance on mathematical typesetting, scientific document formatting, LaTeX workflows, and equation presentation standards for undergraduate, postgraduate, and research-level writing.

An equation with an error in its notation is not a minor formatting issue—it is a scientific error. The difference between and x₂, between sin x and sinx, between a bold vector v and a scalar variable v, carries meaning that readers in every quantitative discipline depend on. Equation formatting is not decoration applied after the mathematics is done—it is part of communicating the mathematics itself. This guide covers the full scope of correct equation presentation: the mechanics of LaTeX and Word equation tools, the conventions that govern how mathematical expressions are displayed in different academic contexts, the discipline-specific standards that determine what correct notation looks like in physics versus chemistry versus statistics, and the common mistakes that appear even in submitted journal papers from experienced researchers.

Why Equation Formatting Is a Scientific Decision, Not an Aesthetic One

Mathematical notation carries meaning through typography. Every formatting choice—whether a variable is italic or upright, whether a symbol is bold or not, whether an operator has spaces around it—communicates something to a reader trained in that discipline. A physicist reading a paper sees the bold upright v and understands immediately that it is a vector quantity with direction, not the scalar speed. A statistician reading N(0,1) knows this is a normal distribution; n without the bracket notation is a sample size. These distinctions are not stylistic preferences—they are the encoding of meaning in the language of mathematics.

90%+ Of peer-reviewed STEM journals require or strongly prefer LaTeX source for papers with equations
ISO 80000 International standard governing mathematical symbol conventions across scientific disciplines
6 types Of LaTeX equation environments covering every display and alignment case a document requires

The international framework for scientific notation is established by ISO 80000, the multi-part standard covering quantities and units across physics, chemistry, mathematics, and engineering. While individual journals and institutions maintain their own style guides, most are consistent with ISO 80000’s core conventions: physical quantities in italic, mathematical operators in upright Roman, vectors in bold italic, tensors in bold upright sans-serif. Understanding this underlying framework explains why the conventions exist rather than presenting them as arbitrary rules to memorise.

The Notation Error That Invalidates the Mathematics

The most consequential equation formatting errors are those that change the meaning of the expression itself: writing a vector without boldface so it appears to be a scalar, omitting parentheses that specify order of operations, conflating the exponential function ex with a generic variable expression, or using the same symbol for two different quantities in the same document. These errors are not caught by spell-check or grammar tools—they survive proofreading by readers who are not expert in the mathematics. They are prevented by deliberate notation decisions made at the start of the document and applied consistently throughout.

The tools available for equation formatting—LaTeX with the American Mathematical Society’s amsmath package, Microsoft Word’s equation editor, MathType, and web-based renderers like MathJax—each handle the typesetting mechanics differently. But all of them require the author to make the same underlying decisions: what type of quantity is each symbol representing, what conventions does this discipline use for that type, and how are those conventions applied consistently when the same quantities appear across dozens of equations throughout a long document.

Inline Equations vs Display Equations: Choosing the Right Mode

Every equation in an academic document exists in one of two placement contexts. Inline equations appear within the running text, embedded in a sentence, formatted to sit within the line height of the surrounding paragraph. Display equations occupy their own horizontal space—centred or left-aligned—separated from the surrounding text by vertical spacing, and are the standard for any expression that will be discussed or referenced elsewhere in the document.

When to Use Inline Mode

Brief variables, simple expressions, and small symbols that introduce a quantity or reference a previously displayed equation. Inline mode suits expressions that can be read without disrupting the sentence flow—single variables like x, brief functions like f(x), or reference to a previously numbered equation. The test: if setting the expression inline makes the line height visibly uneven or forces the reader to pause and decode, it belongs in display mode.

When to Use Display Mode

Complex expressions, derivations, equations with large operators (summation, integral, product), multi-level fractions, matrices, and any equation that will be referred to by number later in the document. Display mode gives the expression room to render at full size without compression—an integral that uses in inline mode shrinks to an inline-sized character, while in display mode it renders at full operator height with limits clearly positioned above and below.

How Inline Math Changes in Different Contexts

LaTeX’s inline math mode uses display-style rendering for fractions, summation limits, and integral limits only when forced with explicit commands. By default, $\frac{a}{b}$ inline produces a compressed fraction set at text height; $\displaystyle\frac{a}{b}$ forces the full-height fraction even within a sentence. The choice matters: an inline fraction that is too visually dominant disrupts the reading flow; one that is too compressed may be illegible. Most style guides and experienced typographers recommend display mode for any expression with fractions, operators with limits, or nested structures.

Inline LaTeX (compressed)
$\frac{1}{n}\sum_{i=1}^{n} x_i$
Renders small — mean formula cramped in text
Display LaTeX (full size)
\[ \frac{1}{n}\sum_{i=1}^{n} x_i \]
Full-height fraction and sum symbol — readable
Appropriate Inline Use
The variable $x_i$ represents…
Single subscripted variable — works inline
Inappropriate Inline Use
The sum $\sum_{i=1}^{n}\frac{x_i^2}{\sigma^2}$
Too complex for inline — needs display mode

LaTeX Math Mode Fundamentals

LaTeX handles mathematical typesetting through dedicated math modes that apply different spacing, font selection, and symbol rendering rules compared to text mode. Every LaTeX math expression must be enclosed within a math-mode delimiter—bare text outside math mode produces text-mode output where symbols like plus signs and letters lose their mathematical meaning and spacing. The choice of delimiter determines the display behaviour: inline or display, numbered or unnumbered.

Math Mode Delimiters

LaTeX Math Delimiters
% ── Inline math modes ──
$expression$               % Standard inline (preferred)
\(expression\)             % Alternative inline (LaTeX preferred syntax)

% ── Display math modes (unnumbered) ──
\[
    expression
\]                         % Display, no number — most common

% ── Display math modes (numbered) ──
\begin{equation}
    expression
    \label{eq:label}
\end{equation}              % Display, auto-numbered, labelled

% ── AVOID (deprecated/broken) ──
% $$expression$$            — eqnarray environment
% Both produce poor spacing compared to amsmath environments

The double-dollar sign delimiter $$...$$ is inherited from plain TeX and behaves differently from LaTeX’s \[...\] in subtle but consequential ways—particularly in vertical spacing above and below the equation. Most LaTeX style guides, including those of the American Mathematical Society and major journals, explicitly recommend avoiding $$...$$ in favour of \[...\] for unnumbered display equations.

Essential Math Mode Commands

Core LaTeX Math Syntax
% Superscripts and subscripts
x^{2}        % Superscript: x²
x_{i}        % Subscript: xᵢ
x^{2}_{i}    % Both: xᵢ²
A_{ij}       % Matrix element

% Fractions
\frac{a}{b}   % Standard fraction a/b
\dfrac{a}{b}  % Display-style fraction (forces full height inline)
\tfrac{a}{b}  % Text-style fraction (forces small size in display)
\cfrac{a}{b}  % Continued fraction (for nested fractions)

% Roots
\sqrt{x+y}          % Square root
\sqrt[3]{x}         % Cube root (nth root with optional argument)

% Named operators (must be upright Roman, not italic)
\sin x    \cos \theta    \tan \alpha
\log n    \ln x          \exp(x)
\lim_{x\to 0}   \max    \min    \det

% Brackets that scale to content
\left( expression \right)
\left[ expression \right]
\left\{ expression \right\}
\left| expression \right|  % Absolute value
\left\| expression \right\| % Norm
The Named Operator Rule

Never write trigonometric functions, logarithms, or other named mathematical operators as plain italic letters in LaTeX. Writing sin x in math mode produces the italicised concatenation sinx—three italic variables multiplied together—not the function sin applied to x. Always use the LaTeX command \sin x, which produces the correct upright Roman format with proper spacing. For operators not built into LaTeX, define custom operators with \DeclareMathOperator{\grad}{grad} in the preamble rather than using \text{grad} inside equations.

The amsmath Package: Why Every Math Document Needs It

The amsmath package, produced by the American Mathematical Society, extends LaTeX’s baseline mathematical capabilities with improved environments, additional operators, and corrected spacing behaviour. It is the authoritative extension for mathematical typesetting in LaTeX and should be loaded in the preamble of every document containing significant mathematical content. The AMS author resources page provides the full documentation alongside style guides for mathematical papers across AMS journals.

What amsmath Adds

  • The align and align* environments for vertically aligned multi-line equations
  • The gather environment for centred multi-line unnumbered equations
  • The multline environment for equations that span more than one line
  • The split environment for splitting a single numbered equation across lines
  • The cases environment for piecewise functions
  • Correct spacing for \left/\right delimiter pairs
  • The \eqref{} command for equation cross-references with parentheses
  • Additional mathematical operators: \operatorname{}, \DeclareMathOperator
  • Improved \text{} within equations for upright Roman text
  • The \tag{} command for custom equation tags

Loading amsmath

Add to your preamble:
\usepackage{amsmath}

For additional symbol sets:
\usepackage{amssymb}

For theorem environments:
\usepackage{amsthm}

The eqnarray environment—present in base LaTeX and frequently appearing in older papers—should be replaced by align from amsmath in all new documents. The spacing produced by eqnarray around the alignment column is non-standard, producing visibly wider spacing around the equals sign than correct mathematical typesetting requires. The amsmath documentation explicitly deprecates eqnarray in favour of the align family.

Equation Numbering: Conventions and Implementation

Equation numbering serves navigation: it gives readers and authors a shared reference point when discussing specific expressions. The decision about which equations to number is both a practical and a stylistic one—numbering every equation in a paper with 60 expressions produces clutter that makes finding referenced equations harder, not easier. Numbering too few means the text must resort to “the equation above” or “the preceding expression”—positional references that break when the document is reformatted.

The Standard Rule for Equation Numbering

Number every displayed equation that is referenced elsewhere in the document. Do not number equations that are displayed for visual clarity alone and never cited.

In LaTeX: use the equation environment (numbered) or \[...\] (unnumbered) based on this rule. If you write a display equation and realise after drafting the surrounding text that you never cite it, switch its environment to \[...\]. If you find yourself writing “the above equation” in the text, switch the referenced equation to a numbered equation environment and use \eqref{}.

Numbering Styles and Sectional Numbering

Numbering StyleAppearanceLaTeX ImplementationTypical Use
Sequential (1), (2), (3) … throughout Default equation environment Short papers, journal articles
Section-based (2.1), (2.2), (3.1) … \numberwithin{equation}{section} in preamble Theses, reports, textbooks with many equations
Chapter-based (1.1), (1.2) … by chapter \numberwithin{equation}{chapter} in preamble PhD theses, monographs
Tagged only Custom label: (Euler), (NS1) \tag{Euler} within equation Named equations referenced by convention
Right-aligned (default) Number at right margin Default in all standard classes Standard for most disciplines
Left-aligned Number at left margin leqno class option Some journals; specific style requirements
Equation Numbering in LaTeX
% Section-based numbering (in preamble)
\numberwithin{equation}{section}

% Standard numbered equation with label
\begin{equation}
    E = mc^{2}
    \label{eq:mass-energy}
\end{equation}

% Referencing (produces "(1)" with parentheses)
As shown in Equation~\eqref{eq:mass-energy}...

% Custom tag for named equations
\begin{equation}
    e^{i\pi} + 1 = 0
    \tag{Euler}
\end{equation}

% Subequations: (2a), (2b), (2c)
\begin{subequations}
    \begin{align}
        a &= b + c  \label{eq:sub-a} \\
        d &= e - f  \label{eq:sub-b}
    \end{align}
\end{subequations}

Variable and Symbol Notation: The Typography of Mathematical Meaning

The most pervasive formatting errors in academic equations involve incorrect type style for variables, constants, and operators. The rules are not arbitrary—they encode the category of each mathematical object: italic for variables, upright for constants and operators, bold for vectors and matrices. Applying these rules consistently is what allows equations to be read correctly by anyone trained in the field, regardless of the language of the surrounding text.

Scalar Variables

Single-letter Roman or Greek variables representing scalar quantities: x, y, t, n, α, β, θ. Set in italic in all math modes. LaTeX applies this automatically to single letters in math mode.

Vectors

Quantities with direction and magnitude: v, F, B. Set in bold italic. LaTeX: \boldsymbol{v} or \bm{v} (with bm package). Some disciplines use arrow notation: \vec{v}.

Matrices and Tensors

Arrays of scalar entries: A, M. Set in bold upright Roman. LaTeX: \mathbf{A}. Tensors in bold upright sans-serif: \mathsf{T}. Second-order tensors may use blackboard bold in some notations.

Named Functions

sin, cos, tan, log, exp, det, tr, div, grad, curl. Always upright Roman—never italic. LaTeX: use built-in commands \sin, \log, or define custom operators.

Mathematical Constants

e (Euler’s number), i (imaginary unit), π. ISO 80000 recommends upright Roman for these: e, i. Practice varies by field—check your target journal’s style guide. LaTeX: \mathrm{e}, \mathrm{i}.

Text Within Equations

Descriptive text embedded in an equation, such as subscript labels: Vmax, Ekinetic. Multi-word subscripts must use \text{} or \mathrm{}—not plain letters, which are treated as multiplied variables.

Type Style in LaTeX Math Mode
% CORRECT: italic variable, upright operator
The velocity \boldsymbol{v} has magnitude $v = |\boldsymbol{v}|$

% CORRECT: multi-word subscript using \text{}
$E_{\text{kinetic}} = \frac{1}{2}mv^2$

% WRONG: multi-word subscript without \text{}
% $E_{kinetic}$  — produces E_{k·i·n·e·t·i·c} (multiplied italic vars)

% CORRECT: upright constants (ISO 80000)
$y = A\mathrm{e}^{-\mathrm{i}\omega t}$

% CORRECT: matrix bold upright
$\mathbf{A}\boldsymbol{x} = \boldsymbol{b}$

% CORRECT: blackboard bold for number sets
$x \in \mathbb{R}, \quad n \in \mathbb{N}, \quad z \in \mathbb{C}$

Common Mathematical Constructs and Their Correct Formatting

Certain mathematical structures appear across disciplines and document types with sufficient frequency that their formatting conventions are worth treating explicitly. These are not edge cases—they are the fraction, integral, summation, limit, and derivative constructs that appear on nearly every page of a quantitative academic paper, and small formatting errors in them are among the most visible issues in submitted manuscripts.

Summation, Integration, and Product Notation

Large Operators — Display vs Inline Behaviour
% Sum with limits — display mode (limits above/below)
\begin{equation}
    S = \sum_{i=1}^{n} x_i
\end{equation}

% Integral with limits
\begin{equation}
    I = \int_{0}^{\infty} f(x)\, \mathrm{d}x
\end{equation}
% Note: \, adds thin space before dx; \mathrm{d} makes differential upright

% Double and triple integrals
$\iint_S \mathbf{F}\cdot\mathrm{d}\mathbf{S}$
$\iiint_V f\, \mathrm{d}V$

% Product notation
\begin{equation}
    P = \prod_{k=1}^{n} a_k
\end{equation}

% Limits
\begin{equation}
    L = \lim_{x \to 0} \frac{\sin x}{x} = 1
\end{equation}
S = Σ xᵢ (display sum)
I = ∫₀^∞ f(x) dx (display integral)

Derivatives and Differential Notation

Derivative notation has consistent conventions that are frequently misapplied. The differential operator d is set upright (Roman) in ISO 80000 notation, not italic—distinguishing it from the variable d that might appear in the same expression. Partial derivatives use , which LaTeX provides as \partial.

Ordinary Derivatives

$\frac{\mathrm{d}y}{\mathrm{d}x}$  % Leibniz notation
$f'(x)$               % Prime notation
$\dot{x}$             % Newton dot (time deriv)
$\frac{\mathrm{d}^2y}{\mathrm{d}x^2}$ % Second order

Partial Derivatives

$\frac{\partial f}{\partial x}$
$\frac{\partial^2 f}{\partial x^2}$
$\frac{\partial^2 f}{\partial x\,\partial y}$
$\nabla f$  % Gradient
$\nabla^2 f$  % Laplacian

Multi-Line and Aligned Equations

Derivations in academic papers regularly require equations to span multiple lines, either because a single expression is too long for one line or because a multi-step derivation needs each step shown. The amsmath package provides distinct environments for each case, each producing different visual output and numbering behaviour.

1 align — Multiple Equations with Vertical Alignment

The align environment is the workhorse for multi-step derivations. Each line receives its own equation number unless suppressed with \nonumber. Lines are aligned at the & symbol—typically placed just before the equals sign or the operator being aligned.

\begin{align}
    f(x) &= (x+1)(x+2)                 \label{eq:factored} \\
         &= x^2 + 3x + 2                   \label{eq:expanded} \\
         &= x^2 + 3x + 2 \nonumber
\end{align}
% align* suppresses all numbers
2 multline — Long Single Equations

Use multline when a single equation must break across lines but is conceptually one expression—not a derivation with multiple steps. The first line is left-aligned, the last line is right-aligned, and intermediate lines are centred. One equation number is produced for the whole expression.

\begin{multline}
    F(x,y,z) = a_{11}x^2 + a_{12}xy + a_{13}xz \\
             + a_{21}yx + a_{22}y^2 + a_{23}yz \\
             + a_{31}zx + a_{32}zy + a_{33}z^2
    \label{eq:quadratic-form}
\end{multline}
3 cases — Piecewise Functions

The cases environment (from amsmath) formats piecewise-defined functions with an automatic large brace on the left and aligned conditions on the right. Each case is a row with the expression and its condition separated by &.

\begin{equation}
    f(x) = 
    \begin{cases}
        x^2       & \text{if } x \geq 0 \\
        -x^2      & \text{if } x < 0
    \end{cases}
\end{equation}
4 gather — Centred Lines Without Column Alignment

The gather environment centres each equation independently without aligning columns. Useful for a series of equations that share a context but have no natural alignment point—each is its own expression rather than a step in a derivation. Each line receives a number; gather* suppresses numbers.

Equations in Microsoft Word: Tools, Techniques, and Limitations

Microsoft Word’s equation editor—introduced in its current form in Word 2007 and progressively improved through Office 365—provides mathematically capable equation insertion for documents that must remain in .docx format. It uses a markup language called UnicodeMath that converts symbolic syntax to rendered output. While Word’s equation rendering is substantially better than early versions, it remains an inferior option to LaTeX for documents with heavy mathematical content, complex derivations, or journal submissions that accept LaTeX source.

Word Equation Editor Limitations

  • Equation numbering requires manual management or workarounds (tables or macros)
  • Cross-references to equation numbers can break on reformatting
  • No automatic sequential numbering like LaTeX equation environment
  • Limited support for multi-line aligned derivations
  • Spacing around operators is less controlled than TeX engine output
  • Converting to PDF may alter equation rendering slightly

Word Equation Editor Advantages

  • No installation required — built into Office
  • Visual toolbar for users unfamiliar with markup syntax
  • Integrates with the surrounding text without compilation step
  • Accessible to non-LaTeX users collaborating on documents
  • MathType compatibility for advanced formatting
  • Accepted in many humanities and social science contexts

The Equation Numbering Workaround in Word

Word has no native automatic sequential equation numbering equivalent to LaTeX’s equation environment. The standard workaround uses a three-column table with no visible borders: the left cell is empty (or contains a tag for subequation labelling), the middle cell contains the equation centred, and the right cell contains the number right-aligned in parentheses. Alternatively, use a tab stop at the right margin with a right-aligned tab character followed by the number. Neither approach is fully automatic—you must update numbers manually if equations are reordered—which is the practical argument for LaTeX in any document with more than a handful of referenced equations.

Word Equation Markup (UnicodeMath / Linear Format)
Keyboard shortcut: Alt+= (Windows) or Insert > Equation

UnicodeMath syntax (type and press Enter or Space to render):
\frac{a}{b}          → fraction a/b
\sqrt{x+y}           → square root of (x+y)
x^2                  → x superscript 2
x_i                  → x subscript i
\int_0^\infty f(x)dx → integral from 0 to ∞
\sum_{i=1}^n x_i    → summation
\alpha \beta \gamma  → Greek letters α β γ
\partial f/\partial x → partial derivative

Named operators: \sin \cos \tan \log \ln \exp \lim
Bold: use Format > Bold after selecting within equation
Upright Roman: use \mathrm{} notation

MathType for Word

For Word users producing documents with substantial mathematical content, MathType (now incorporated into Wiris) extends Word’s built-in equation editor with proper LaTeX input, automatic equation numbering, more symbol support, and better cross-reference management. It also exports equations in MathML for web documents. For documents that must be in Word format but contain LaTeX-quality mathematics, MathType closes much of the gap between Word and native LaTeX output. Check whether your target journal accepts Word submissions with MathType equations, as some LaTeX-based submission portals may not parse them correctly.

Discipline-Specific Equation Formatting Standards

Mathematical notation is not uniform across academic disciplines. Physicists, statisticians, engineers, chemists, and economists each use conventions that are internally consistent but differ across fields. A notation that is standard in one discipline may be ambiguous or incorrect in another. Understanding which conventions govern your specific field prevents the category of errors that are invisible to spell-checkers but immediately apparent to expert readers.

Pure Mathematics

Sets in blackboard bold (ℝ, ℂ, ℕ, ℤ). Operators upright. Proofs end with ∎ (tombstone) or QED. Theorem and proof environments. Vectors often as elements of abstract spaces without bold notation.

Physics

ISO 80000 conventions strongly followed. Differential d upright. Constants e, i upright where possible. SI units in upright Roman separated from quantity by thin space. Bra-ket notation: ⟨ψ|Ĥ|ψ⟩ (use braket package in LaTeX).

Statistics

Random variables in capital italic: X, Y, Z. Realised values in lowercase: x, y, z. Probability P(·), expectation 𝔼[·] or E[·]. Distributions: X ~ N(μ, σ²). Hat notation for estimates: β̂.

Chemistry

Element symbols upright Roman, never italic. Subscripts for atom count, superscripts for charge: H₂O, Na⁺, SO₄²⁻. Reaction arrows: ⟶, ⇌. Use mhchem package in LaTeX. Isotopes: 12C, 14N.

Engineering

Units always upright Roman, separated from value: 9.81 m s⁻². Vectors bold italic, matrices bold upright. Signal notation: x(t) time domain, X(f) or X(ω) frequency domain. Transfer functions: H(s), H(z).

Economics

Utility functions: U(x). Parameter vectors bold: β. Expectation: 𝔼[·] or E[·]. Optimal values with asterisk: x*. Lagrange multiplier λ upright in applied work. Matrices in bold upright for econometric models.

The safest discipline-specific resource is always the author instructions of your target journal or the official style manual of the professional society in your field. The AMS publishes extensive author guidelines for mathematical papers; the American Physical Society (APS) Physical Review journals have detailed notation guides; the American Chemical Society has explicit style requirements for chemical equations. When in doubt, examine how equations are formatted in recent published papers in the specific journal you are targeting—consistent published examples outrank general rules when they differ.

Chemical Equation Formatting

Chemical equations follow conventions that differ substantially from mathematical equations. The critical distinction is that element symbols are not variables—they represent specific elements and must be set in upright Roman type, never italic. A hydrogen atom is H (upright), not H (italic); the latter would be interpreted as a variable in a mathematical expression.

The mhchem Package for LaTeX Chemical Equations

The mhchem package provides the \ce{} command, which handles all chemical equation formatting automatically: upright element symbols, correct subscripts for atom counts, superscripts for ionic charges, reaction arrows, equilibrium arrows, state symbols, and isotope notation. Load it with \usepackage[version=4]{mhchem}. The package correctly interprets chemical notation that would otherwise require manual construction from multiple commands, and it is the standard approach in chemistry and biochemistry documents produced in LaTeX.

Chemical Equations with mhchem
\usepackage[version=4]{mhchem}  % In preamble

% Basic reaction
\ce{2H2 + O2 -> 2H2O}

% With state symbols
\ce{NaCl(s) + H2O(l) -> Na+(aq) + Cl-(aq)}

% Equilibrium reaction
\ce{CO2 + H2O <=> H+ + HCO3-}

% Isotopes
\ce{^{14}_{6}C -> ^{14}_{7}N + e- + \bar{nu}_e}

% Ionic charges and complex formulae
\ce{Fe^{2+} + 2OH- -> Fe(OH)2}
\ce{SO4^{2-}}
\ce{[Cu(NH3)4]^{2+}}  % Coordination complex

Chemical equations in Word or non-LaTeX contexts require manual attention to the same conventions. Subscripts in molecular formulae are true subscripts (2), not variables—use the subscript formatting button in Word rather than underscore characters. Charge superscripts follow the number before the sign: 2+ not +2 (the conventional order is magnitude then sign). Reaction arrows should be proper arrow characters: → (U+2192) for one-way reactions, ⇌ (U+21CC) for equilibrium—not a hyphen-greater-than combination.

Matrices and Arrays

Matrix notation is one of the areas where LaTeX has the clearest advantage over Word: matrix construction in LaTeX is systematic and scales cleanly to large arrays, while equivalent constructs in Word require careful manual formatting that becomes difficult to maintain as matrices grow. The core LaTeX environment is pmatrix (parenthesis brackets), with variants for different bracket styles.

Matrix Environments in LaTeX
\usepackage{amsmath}  % All matrix envs require amsmath

% pmatrix: parentheses ( )
\begin{pmatrix}
    a & b & c \\
    d & e & f \\
    g & h & i
\end{pmatrix}

% bmatrix: square brackets [ ]
\begin{bmatrix} a & b \\ c & d \end{bmatrix}

% vmatrix: vertical bars | |  (determinant)
\begin{vmatrix} a & b \\ c & d \end{vmatrix}

% Vmatrix: double vertical bars ‖ ‖  (norm)
\begin{Vmatrix} a & b \\ c & d \end{Vmatrix}

% Large matrix with dots for implied entries
\begin{pmatrix}
    a_{11} & a_{12} & \cdots & a_{1n} \\
    a_{21} & a_{22} & \cdots & a_{2n} \\
    \vdots & \vdots & \ddots & \vdots \\
    a_{m1} & a_{m2} & \cdots & a_{mn}
\end{pmatrix}

The \cdots, \vdots, and \ddots commands produce horizontal, vertical, and diagonal ellipsis dots used in matrices where full entries are implied by pattern. These are mathematically meaningful symbols—\cdots indicates a horizontal sequence of omitted elements aligned with the row; \vdots indicates a vertical sequence aligned with the column; \ddots indicates a diagonal sequence. Using plain ... (ellipsis in text mode) inside a matrix is incorrect—it produces text-mode dots with text spacing rather than mathematically correct centred dots.

For augmented matrices—used in linear algebra to show the coefficient matrix and constant vector side by side in row reduction—the base amsmath matrix environments need the column separating line added manually using array with a custom column specification: \left(\begin{array}{ccc|c} ... \end{array}\right). The array environment is the general matrix construction tool; pmatrix, bmatrix, and the others are convenience wrappers around it with fixed bracket styles.

Referencing Equations in Text

How an equation is cited in the surrounding prose affects both readability and correctness. The reference form must be consistent throughout the document, and the LaTeX implementation should use \eqref{} rather than \ref{} to ensure parentheses are automatically included and correctly formatted.

Correct Reference Forms

  • “Equation (3) shows that…” — full word at sentence start
  • “…as given by Eq. (3)” — abbreviated form mid-sentence
  • “…the result (3) implies…” — numeral only when context is clear
  • “From Equations (2) and (3)…” — plural form
  • “…combining Eqs. (4)–(6)…” — range with en-dash

Incorrect Reference Forms

  • “The above equation shows…” — positional, breaks on reformat
  • “The following equation shows…” — positional
  • “equation 3” — equation not capitalised; number not in parens
  • “Eqn. (3)” — non-standard abbreviation
  • “eq. 3” — lowercase and no parentheses
LaTeX Equation Cross-References
% \eqref{} produces (1) with parentheses — use this for equations
Substituting into Equation~\eqref{eq:momentum} gives...

% \ref{} produces 1 without parentheses — correct for sections, figures
% but not standard for equations

% The tilde ~ prevents line break between "Equation" and "(1)"
% Always use ~ before \eqref and \ref

% Referring to a range of equations
From Equations~\eqref{eq:first}--\eqref{eq:last}...

Web-Based Equation Rendering: MathJax and MathML

Academic content published on the web—HTML journal articles, preprint servers, academic blogs, online course materials—requires a rendering mechanism for equations that works within browsers without requiring PDF conversion. Two technologies govern this space: MathML (Mathematical Markup Language), the W3C standard for encoding mathematical expressions in HTML, and MathJax, the JavaScript library that converts LaTeX or MathML syntax to rendered output in any browser.

MathJax: LaTeX in the Browser

MathJax enables academic authors to write LaTeX equation syntax in HTML documents and have it rendered correctly in any modern browser without any browser-side installation. The library is loaded from a CDN and intercepts LaTeX delimiters ($...$ and \[...\]) in the page source, converting them to SVG or MathML output that matches LaTeX-quality typesetting. Most major preprint servers—arXiv, bioRxiv—and many online journals use MathJax as their equation rendering layer, which is why the LaTeX syntax authors write in their source files renders correctly in both the PDF and the HTML versions of their paper.

For web documents where you need equations: include MathJax via its CDN link, write LaTeX syntax in your HTML, and MathJax renders it automatically. The output is accessible to screen readers when the MathML accessibility extensions are enabled—an advantage over equation images, which carry no mathematical semantic information.

MathML itself is XML-based markup that browsers can theoretically render natively, but browser support has historically been inconsistent—Firefox has long supported MathML natively; Chrome added native MathML Core support in 2023. For production web equations, MathJax remains the more reliable approach because it does not depend on browser-specific MathML implementation quality.

Equation Accessibility in Academic Documents

Equations in PDF documents are, by default, inaccessible to screen readers and assistive technologies used by readers with visual impairments or reading disabilities. A sighted reader interprets the visual typesetting of an equation; a screen reader attempting to read a PDF equation rendered as an image or typeset characters reads either nothing or a garbled character sequence that conveys no mathematical meaning. Accessibility in equation formatting is not an edge case—it affects a significant proportion of the academic readership and is increasingly required by funding bodies, universities, and publishers.

Making LaTeX Equations Accessible

LaTeX PDFs with equations can be made accessible through several mechanisms. The accessible package adds alternate text to mathematical content. The axessibility package (from the LaTeX accessibility project) structures mathematical content for screen reader output. For HTML output from LaTeX, make4ht and tex4ht convert LaTeX to HTML with MathML output, producing the most fully accessible mathematical web content. For new documents, producing both a PDF and an HTML version with MathJax rendering provides the broadest accessibility coverage.

The Overleaf documentation provides detailed guidance on mathematical expressions and accessibility for LaTeX documents, covering both the correct environments and the accessibility considerations for different output targets.

Alt Text for Equation Images

In Word documents, PowerPoint presentations, or web pages where equations are inserted as images (a common workflow for non-LaTeX users copying from equation editors), each image must have alt text that describes the mathematical expression. Alt text for equations should be the full spoken description of the equation: “E equals m c squared” rather than “equation 1” or the empty alt text that is the default for inserted images. Many institutional and publisher accessibility requirements now mandate equation alt text, and automated accessibility checkers flag equation images without it.

Journal and Institution Submission Requirements

The specific equation formatting requirements of academic journals and institutional submission portals take precedence over any general convention. Different publishers, different journals within the same publisher, and different disciplinary traditions all have requirements that may diverge from the conventions described in this guide. The author instructions page of your target journal is the definitive reference—reading it before beginning to write, not after the first draft is complete, saves substantial reformatting effort.

Before Writing

Read the Author Instructions

Identify: Does the journal require LaTeX or accept Word? Which document class or template should you use? Is there a specific equation numbering style? Are there constraints on equation placement, size, or font? Download the official journal template if one exists—these encode all formatting requirements automatically.

During Writing

Use the Official Template or Class File

Major publishers provide LaTeX class files (.cls) that handle equation numbering, font, spacing, and layout automatically. Loading the correct class file removes the need to manually configure most formatting decisions. For AMS journals, use the amsart class; for Elsevier, elsarticle; for IEEE, ieeetran. Templates are available in the Overleaf template gallery.

Before Submission

Verify Equation Formatting Against the Style Guide

Check: Are all referenced equations numbered? Is the numbering style correct (sequential, section-based)? Are vectors, matrices, and constants formatted consistently with the journal’s notation guide? Do all named operators use upright Roman type? Are chemical equations (if applicable) formatted with correct element notation?

At Submission

Submit Source Files, Not Just PDFs

Most LaTeX-based journals require submission of the .tex source file, all .bib and custom style files, and all figures—not just the compiled PDF. The editorial system re-compiles your source to verify it compiles cleanly and to allow typesetting staff to make any final adjustments. Ensure your source compiles without errors before submission.

AMS Uses amsart class; AMS-LaTeX conventions; equation numbers right-aligned
IEEE IEEEtran class; numbered equations; specific vector/matrix notation guide
Elsevier elsarticle class; accepts both LaTeX and Word; journal-specific style variations
Springer svjour3 or springer-mathphys style; author check list for notation

Common Equation Formatting Mistakes in Academic Documents

The following errors appear consistently in student submissions, thesis drafts, and even submitted journal manuscripts. Most are invisible to non-specialist proofreaders, which is exactly what makes them persistent. Recognising each pattern allows you to audit your own documents deliberately rather than relying on review to catch them.

Italic Named Functions

Writing sin x, log n, or exp(x) in math mode without the LaTeX operator commands produces italicised letter sequences that parse as products of variables, not function applications. A reader sees s × i × n × x — four variables multiplied — rather than the sine function applied to x. Fix: Always use \sin, \log, \exp, \det, etc. For custom operators, declare them with \DeclareMathOperator in the preamble.

Unscaled Brackets Around Large Expressions

Writing (\frac{a}{b}) produces parentheses at text height around a full-height fraction — the brackets are visually mismatched and difficult to parse. Fix: Use \left( \frac{a}{b} \right) for brackets that automatically scale to the height of their contents. This applies to all bracket types: parentheses, square brackets, curly braces, absolute value bars, and norms.

Multi-Word Subscripts Without \text{}

Writing E_{kinetic} produces Ek·i·n·e·t·i·c — each letter treated as a separate italic variable in a product. Fix: Use E_{\text{kinetic}} or E_{\mathrm{kinetic}} to render the subscript as upright Roman text. The \text{} command uses the current font; \mathrm{} always uses upright Roman regardless of context.

The Wrong Minus Sign

There are multiple typographically distinct horizontal line characters: the hyphen (-), the en-dash (–), the em-dash (—), and the mathematical minus sign (−). In equations, only the mathematical minus sign is correct, and it only appears in math mode. Writing a temperature like −5°C requires the math-mode minus: $-5$°C or the Unicode minus sign U+2212. Using a hyphen in place of a minus sign in a displayed equation is a visible typographic error that signals unfamiliarity with mathematical typesetting conventions.

Inconsistent Notation for the Same Quantity

Using x in some equations and X in others for the same quantity, switching between bold and arrow vector notation mid-document, or defining a variable in Section 1 and reusing the same symbol for a different variable in Section 4 without redefinition—these are not minor oversights. They produce genuine mathematical ambiguity that reviewers flag as errors. Fix: Maintain a notation table in your drafting process listing every symbol, what it represents, its type style, and where it is first defined. Review the whole document before submission against this table.

Equations as Images in Web or Word Documents

Inserting equations as screenshots or exported image files produces pixelated output that degrades at different zoom levels, cannot be copied or searched, and is entirely inaccessible to assistive technologies. Fix: For Word documents, use the built-in equation editor or MathType. For web content, use MathJax or a LaTeX-to-HTML workflow. For PDFs, use LaTeX. Equation images are never the correct choice for any academic document type.

The Proofreading Checklist for Equations
  • All named operators use upright Roman commands
  • All brackets scale with \left/\right
  • All multi-word subscripts use \text{}
  • Vectors and matrices consistent notation throughout
  • All referenced equations are numbered
  • All equation numbers cited as (n), not eq. n
  • No positional references (“the above equation”)
  • Differential d is upright where required
  • Chemical element symbols are upright
  • Units are upright, separated from values
  • No equation images — all typeset natively
  • Notation consistent with target journal style

Need Help with Mathematics-Intensive Academic Documents?

From LaTeX typesetting of complex derivations to formatting equations correctly for specific journal submission requirements, our academic specialists provide support across all quantitative disciplines.

Greek Letters, Special Symbols, and Spacing in Equations

Greek letters and special mathematical symbols are the vocabulary of quantitative disciplines—every student of mathematics, physics, chemistry, or engineering spends years developing fluency with them. LaTeX provides commands for every Greek letter and thousands of additional symbols; using the correct command rather than copy-pasting a Unicode character from elsewhere ensures that symbols render with correct mathematical spacing, scaling, and PDF embedding.

Greek Letters Reference

Common Greek Letters in LaTeX Math Mode
% Lowercase Greek
\alpha   α    \beta    β    \gamma   γ    \delta   δ
\epsilon ε    \varepsilon ε  \zeta    ζ    \eta     η
\theta   θ    \vartheta   ϑ  \iota    ι    \kappa   κ
\lambda  λ    \mu      μ    \nu      ν    \xi      ξ
\pi      π    \varpi   ϖ    \rho     ρ    \sigma   σ
\tau     τ    \upsilon υ    \phi     φ    \varphi  φ
\chi     χ    \psi     ψ    \omega   ω

% Uppercase Greek (those that differ from Roman letters)
\Gamma   Γ    \Delta   Δ    \Theta   Θ    \Lambda  Λ
\Xi      Ξ    \Pi      Π    \Sigma   Σ    \Upsilon Υ
\Phi     Φ    \Psi     Ψ    \Omega   Ω

Mathematical Spacing Commands

LaTeX’s automatic spacing in math mode handles most situations correctly, but certain constructs require manual spacing adjustments. The most common: a thin space before the differential in integrals, a thin space to separate a sum symbol from its summand when they would otherwise appear cramped, and negative space to bring together elements that LaTeX separates too much.

Thin space (most common)
\int f(x)\,\mathrm{d}x
Adds small gap before dx — standard practice
Medium space
f(x)\: g(x)
Slightly wider gap between adjacent functions
Quad space
x \quad \text{where} \quad x > 0
Large space separating equation from condition
Negative thin space
\!\sqrt{x}
Reduces space — corrects over-spacing in rare cases

Theorem, Proof, and Definition Environments

Mathematics and related disciplines use structured environments—theorems, lemmas, propositions, corollaries, definitions, proofs, remarks—that give formal status to specific results and arguments. These environments are provided by the amsthm package and are standard in mathematical writing from undergraduate problem sets through to research monographs. Their correct use signals familiarity with mathematical writing conventions; their absence or misuse signals the opposite.

Theorem-Style Environments (amsthm package)
\usepackage{amsthm}

% Define theorem-like environments in preamble
\newtheorem{theorem}{Theorem}[section]
\newtheorem{lemma}[theorem]{Lemma}
\newtheorem{corollary}[theorem]{Corollary}
\newtheorem{proposition}[theorem]{Proposition}
\theoremstyle{definition}
\newtheorem{definition}[theorem]{Definition}
\theoremstyle{remark}
\newtheorem{remark}[theorem]{Remark}

% Usage in document
\begin{theorem}[Pythagoras]\label{thm:pythagoras}
    For a right triangle with legs $a$, $b$ and hypotenuse $c$,
    $a^2 + b^2 = c^2$.
\end{theorem}

\begin{proof}
    Consider a square with side $a + b$...
\end{proof}
% \end{proof} automatically adds the QED tombstone ∎

The three theorem styles available in amsthm govern the visual presentation: plain (bold header, italic body — for theorems, lemmas, propositions), definition (bold header, upright body — for definitions and examples), and remark (italic header, upright body — for remarks and notes). Choosing the right style for each environment type is part of mathematical document formatting convention, not just visual preference.

Integrating Equations into Academic Prose

The relationship between equations and the surrounding text is one of the most commonly mishandled aspects of academic writing in quantitative disciplines. Equations are not ornaments appended to paragraphs—they are statements that must be grammatically and logically integrated into the prose. A displayed equation should be able to be read aloud as part of the sentence it belongs to, with the appropriate grammatical connectives in place.

“Every equation is a sentence—or part of one. It should be punctuated accordingly and connected to what precedes and follows it with the same care as any other statement in the argument.”

Good Integration

The total kinetic energy of the system is

T = ½mv², (4)

where m is the mass and v the speed. Note the comma after the equation and the grammatical continuation.

Poor Integration

Kinetic energy is shown in the below equation:

T = ½mv²

The equation is unnumbered, preceded by a colon that disconnects it from the prose, and followed by a new paragraph that begins without explanation of what the variables mean.

Punctuation of display equations follows the same rules as punctuation of any sentence component: if the equation ends the sentence, it takes a period; if the sentence continues after the equation, the equation takes a comma; if the equation is in the middle of a clause, no punctuation may be needed. This is not universally observed in all journals and disciplines—some style guides explicitly forbid equation punctuation—but the underlying principle of grammatical integration applies everywhere: equations are not separate from the text, they are part of it.

Variables introduced in equations must be defined on their first appearance, either immediately before the equation (“where x is the displacement”) or immediately after (“where α is the decay constant”). Undefined variables in equations create ambiguity that peer reviewers flag regardless of how correct the mathematics is. For documents with many variables, a nomenclature section listing all symbols with their definitions, types, and units provides readers with a reference table—standard in engineering theses and long technical reports.

For students and researchers who need expert assistance producing mathematically rigorous written work—whether a dissertation with derivations, a lab report with equations, or a journal submission requiring LaTeX typesetting to a specific style—our science writing service and mathematics assignment support provide specialist guidance across all equation-intensive academic contexts. For documents that need professional review before submission, our proofreading and editing service includes equation notation checking as part of technical document review.

FAQs About Equation Formatting

Should equations be numbered in academic papers?
Number every displayed equation that you reference later in the text. Equations shown for illustration only and never cited can be left unnumbered using the \[...\] environment in LaTeX or the starred equation* environment. Over-numbering clutters the document and is considered poor style. The standard form for numbered equations is parenthetical Arabic numerals at the right margin: (1), (2), (3). In theses and longer reports, section-based numbering—(2.1), (2.2)—helps readers locate specific equations in long documents.
What is the difference between inline and display equations?
Inline equations appear within the flow of a sentence, formatted at text line height. Display equations occupy their own centred line with vertical spacing above and below. The rule of thumb: any expression complex enough to be cramped or difficult to read in inline mode—fractions, integrals, summations with limits, matrices—belongs in display mode. Simple variables, short expressions, and references to previously numbered equations are appropriate inline. The choice affects both readability and the visual rhythm of the page.
How do I format equations in Microsoft Word?
Insert an equation using Alt+= (Windows) or Insert → Equation (Mac). Word uses UnicodeMath syntax: type \frac{a}{b} for fractions, \sqrt{x} for square roots, x^2 for superscripts. For numbered display equations, use a three-column borderless table with the equation centred in the middle cell and the number right-aligned in the right cell. For documents with more than a few referenced equations, MathType provides automatic sequential numbering and proper cross-references that Word’s built-in editor lacks.
Which LaTeX environment should I use for multi-line equations?
For vertically aligned multi-step derivations, use the align environment from amsmath—place & before the alignment point (usually the equals sign) and \\ between lines. For equations too long for one line that are one logical expression, use multline. For piecewise functions, use cases. Avoid the outdated eqnarray environment—it has known spacing defects that amsmath environments do not share. Use align*, gather*, or multline* for the unnumbered versions of each environment.
How should variables be formatted in equations?
Scalar variables in italic: x, y, t, n, α, β. Vectors in bold italic: v, F, B — use \boldsymbol{v} or \bm{v} in LaTeX. Matrices in bold upright Roman: use \mathbf{A}. Named functions (sin, cos, log, exp) in upright Roman — use the LaTeX built-in commands \sin, \log, etc. Multi-word subscripts in upright text — use \text{kinetic} inside math mode. Mathematical constants e, i, π in upright Roman per ISO 80000, though practice varies by discipline.
How do I write chemical equations with correct formatting?
In LaTeX, load the mhchem package and use the \ce{} command: \ce{2H2 + O2 -> 2H2O}. This automatically handles upright element symbols, subscripts, superscripts for charges, reaction arrows, state symbols, and isotope notation. In Word, use the equation editor with upright Roman for element symbols (not italic), true subscripts for atom counts, superscripts for charges in the correct order (magnitude then sign: 2+ not +2), and proper arrow characters (→, ⇌) rather than typed substitutes.
What is MathML and when should I use it?
MathML is the W3C standard for encoding mathematical notation in web documents. It is the output format used by MathJax, which converts LaTeX syntax to browser-renderable mathematics. For academic PDFs, MathML is not directly relevant—use LaTeX. For web-based publishing, HTML academic documents, or content requiring screen-reader accessibility, MathML ensures equations are parseable by assistive technologies rather than rendered as inaccessible images. Most authors who need web equations write in LaTeX syntax and use MathJax for rendering, which handles the MathML conversion automatically.
How do I reference an equation in the text?
Use “Equation (n)” or “Eq. (n)” — “Equation” in full at the start of a sentence, abbreviated “Eq.” mid-sentence. In LaTeX, use \eqref{label} from amsmath, which automatically produces the number in parentheses: “…as shown in Equation~\eqref{eq:label}.” Never use positional references like “the above equation” or “the following expression” — these break when equations are renumbered or when the text layout shifts. Always use the equation number as the reference anchor.

From Notation Decisions to Submission-Ready Documents

Equation formatting is a sequence of decisions, not a single task performed at the end of writing. The decision about which tool to use—LaTeX or Word—shapes every subsequent formatting choice. The decision about notation conventions—bold or arrow vectors, upright or italic differential, sequential or sectional numbering—shapes the document’s internal consistency from first equation to last. The decision about which environments to use for multi-line expressions determines whether derivations are readable or cramped.

Making these decisions explicitly at the start of a document, documenting them in a notation list, and applying them consistently throughout produces the kind of equation formatting that reviewers do not comment on—because correctly formatted equations are invisible as formatting and visible only as mathematics. The goal of equation formatting is to remove itself from the reader’s attention entirely, leaving only the mathematical content.

For quantitative academic work where equation formatting must meet the specific standards of a journal, institution, or examination board, the resources available are extensive: the AMS LaTeX author guide, the Overleaf documentation on mathematical expressions, the style guides of individual journals, and the accumulated conventions of your specific discipline all provide authoritative reference points. The investment in learning them pays compound returns across every paper, thesis, and technical document you produce.

Related Academic Writing and Formatting Resources

For students working with LaTeX across full academic documents, our guides on dissertation and thesis writing and citation and referencing cover the broader document-level context for mathematical work. For assignment-level support in quantitative subjects, explore our statistics assignment help, physics and geometry support, calculus homework help, and data analysis assistance. For formatting support on complete papers, our paper formatting service covers both equation notation and broader document compliance with institutional and journal standards.

Expert Support for Scientific and Mathematical Writing

Our specialists assist with LaTeX typesetting, equation formatting, and the full range of quantitative academic document production.

Get Writing Support
Article Reviewed by

Simon

Experienced content lead, SEO specialist, and educator with a strong background in social sciences and economics.

Bio Profile

To top