Add equation name underneath equation number

Question

I am wondering if it is possible (and if so how) to add names to equations just underneath the equation number. I am just using the standard equation environment to display the equation centered in the middle of the page, with the equation number flushed to the right hand side. I need to put some text immediately below the equation number to serve as the equation's name. Other requirements include:

• The text should be horizontally right-aligned with the equation number and should not affect the latter's vertical alignment with the equation; i.e. the equation and equation number should still be vertically in line with each other.
• The vertical spacing between the equation number and name should be standard text spacing; i.e. if the equation is vertically tall, there should not be excess space between the equation number and name.
• The text should be able to be formatted in different font styles. Right now, I need to format it in \sf.

Here is a MWE (without the equation name):

\documentclass{article}
\usepackage{amsmath}

\numberwithin{equation}{section}

\begin{document}

\section{First Section}
\label{sec:section1}
\begin{equation}
\label{eqn:label}
    \boxed{M\approx\frac{\pi}{4}\left(\frac{2d}{\lambda_o}\right)^2\left(\mathrm{NA}\right)^2}
\end{equation}

Reference \eqref{eqn:label}.

\end{document}

Here is a snapshot example of what I'd like to achieve (taken from Saleh, 2007):

I have tried using the \tag command, but realized that this not only replaces the equation number, which I do not want, but also causes the equation number counter to skip that particular equation. Ideally, the equation should still function as a regular numbered equation, that can be number-referenced (via \eqref or other), but with the addition of some text below the equation number.

Any help will be much appreciated.

Answer 1

Here's an easy by using the starred version of \tag from amsmath:

\documentclass{article}
\usepackage{amsmath}% http://ctan.org/pkg/amsmath
\newcommand{\eqname}[1]{\tag*{#1}}% Tag equation with name
\begin{document}

\begin{align}
  f(x) &= a \\ \eqname{Constant} \\
  g(x) &= ax \\ \eqname{Linear} \\
  h(x) &= ax^2+bx+c \label{abc} \\ \eqname{Quadratic}
\end{align}
See~\eqref{abc}.
\end{document}

For consistency, I've wrapped \tag* inside \eqname.

---

Depending on your equation construction, you can adjust the vertical skip between the equation and the name:

\documentclass{article}
\usepackage{amsmath}% http://ctan.org/pkg/amsmath
\newcommand{\eqname}[1]{\tag*{#1}}% Tag equation with name
\begin{document}

\begin{align}
  \boxed{M \approx \frac{\pi}{4}\biggr(\frac{2d}{\lambda_0}\biggr)^2(\text{NA})^2} \label{abc} \\[-\baselineskip]
  \eqname{Number of TE Modes}
\end{align}
See~\eqref{abc}.
\end{document}

Here I've moved it up by \baselineskip, but you can adjust this to your liking.

Answer 2

The following example hacks into the internals of package amsmath to add the equation name at the stage, where the equation tag is set in the displayed equation.

• \eqname{<name of the equation>} sets the equation name. The name increases the depth of the equation tag, thus the base line of the equation number does not change.

• The width of the equation name decreases the space that is available for the equation. Thus the equation number might be moved down (feature of amsmath) as shown in equation 5 of the example.

  The star form \eqname* ignores the width of the equation name. It can be used, if it is clear, that the equation name does not collide with the contents of the equation, see equation 6 of the example.

• Package nameref is supported to reference the name of the equation. The label name is given in the optional argument. (The example uses package nameref, but the package is not required.)

• The formatting of the equation name can be changed by redefining \eqnameref. The example uses \textsf as requested in the comment.

Example:

\documentclass[a5paper]{article}
\usepackage{amsmath}
\usepackage{nameref}

\newcommand*{\eqnameformat}[1]{%
  \textsf{#1}%
}

\makeatletter
\@ifdefinable{\org@maketag@@@}{%
  \let\org@maketag@@@\maketag@@@
  \renewcommand*{\maketag@@@}[1]{%
    \org@maketag@@@{%
      \@ifundefined{eq@name}{#1}{%
        \begin{tabular}[t]{@{}r@{}}%
          #1\tabularnewline
          \eqnameformat{\@nameuse{eq@name}}%
        \end{tabular}%
      }%
    }%
  }%
}
\newif\ifeqname@star
\newcommand*{\eqname}{%
  \@ifstar{\eqname@startrue\eqname@}{\eqname@starfalse\eqname@}%
}
\newcommand*{\eqname@}[2][]{%
  \gdef\eq@name{#1}%
  \ifx\eq@name\@empty
  \else
    \begingroup
      \@ifundefined{GetTitleString}{%
        \gdef\@currenteqlabelname{#2}%
      }{%
        \GetTitleString{#2}%
        \global\let\@currenteqlabelname\GetTitleStringResult
      }%
      \let\@currentlabelname\@currenteqlabelname
      \label{#1}%
    \endgroup
  \fi
  %
  \gdef\eq@name{#2}%
  \ifx\eq@name\@empty
    \global\let\eq@name\relax
  \else
    \ifeqname@star
      \gdef\eq@name{\llap{#2}}%
    \fi
  \fi
}
\@ifdefinable{\org@make@display@tag}{%
  \let\org@make@display@tag\make@display@tag
  \def\make@display@tag{%
    \@ifundefined{@currenteqlabelname}{}{%
      \let\@currentlabelname\@currenteqlabelname
    }%
    \org@make@display@tag
  }%
}
\let\eq@name\relax
\let\@currenteqlabelname\relax
\g@addto@macro\displ@y@{%
  \global\let\eq@name\relax
  \global\let\@currenteqlabelname\relax
}
\@ifdefinable{\org@math@cr@@}{%
  \let\org@math@cr@@\math@cr@@
  \def\math@cr@@[#1]{%
    \org@math@cr@@[{#1}]%
    \noalign{%
      \global\let\eq@name\relax
    }%
  }%
}
\@ifdefinable{\org@eqref}{%
  \let\org@eqref\eqref
  \renewcommand*{\eqref}[1]{%
    \begingroup
      \let\eq@name\relax
      \org@eqref{#1}%
    \endgroup
  }%
}
\g@addto@macro\equation{%
  \eqname{}%
}
\makeatother

\begin{document}

% some test cases

The first equation \eqref{eq:einstein} with  name ``\nameref{eq:einstein}'',
the second equation \eqref{eq:second} and
and the last equation \eqref{eq:last}  with name ``\nameref{eq:last}''.

\begin{gather}
\eqname[eq:einstein]{Einstein}
E=mc^2\\
\label{eq:second}
a=b
\end{gather}

\begin{equation}
\eqname{Name AB}
a=b
\end{equation}

\begin{gather}
x=y
\end{gather}

\begin{gather}
\eqname{A longer equation name A/no star}
a=b=c=d=e=f=g=h=i=j\\
\eqname*[eq:last]{A longer equation name B/star}
a=b=c=d=e=f=g=h=i=j
\end{gather}
\end{document}

Answer 3

Here's a solution. I introduce the environment Nequation to mean "named equation," in which the name is passed as an argument. I show how it does not screw up the use of equation environment's numbering scheme. [EDITED to fix referencing by label, and to add sections to eq #, and also to allow text formatting to name, per user request]

In response to the user's latest query, sub and sub-sub sections can be added where I first \renewcommand\theequation{\thesection.\arabic{equation}}. Text formatting can come in one of two ways. It can be made to happen for every label, as I have done with \small in the Nequation environment, or it can be done on a label-by-label basis, as I show with \textsf in the actual invocation line.

\documentclass{article}
\usepackage{amsmath}
\usepackage{stackengine}
\def\stackalignment{r}
\def\useanchorwidth{T}
\def\stacktype{L}
\newlength\eqshift
\setlength\eqshift{\widthof{)}}
\renewcommand\theequation{\thesection.\arabic{equation}}
\let\savetheequation\theequation
\newenvironment{Nequation}[1]{%
  \def\thecurrentname{#1}%
  \let\theequation\savetheequation%
  \begin{equation}%
  \renewcommand\theequation{%
    \stackunder{\savetheequation}%
    {{\small\thecurrentname}\hspace{-\the\eqshift}}}%
}{%
  \end{equation}%
  \let\theequation\savetheequation%
  \ignorespacesafterend%
}
\begin{document}
\section{New Section}
\begin{Nequation}{\textsf{Equation Name}}
  \label{eqn:label}
    y=mx+c
\end{Nequation}
We find in equation~\eqref{eqn:label}, a different result than
\begin{equation}
    \label{eqtwo}
    y=mx^2+c
\end{equation}
However, in equation~\eqref{eqtwo}\ldots
\end{document}