I guess the main issues are, how to create a large open square symbol, center it on the math axis, and have it act as a math operator, i.e., place its argument below (and above, if there's an upper limit) the large open square. Sort of like \sum, right? A series of \vcenter, \hbox, and repeated \mathlarger instructions applied to the \Box macro solve these issues. (If you compare the following screenshot with the preceding one, you'll notice that the letter g to the left of the box is placed ever so slightly higher now.)
\documentclass{article}
\usepackage{amsmath, % for "\DeclareMathOperator*" macro
amsfonts, % for "\Box" macro
relsize} % for "\mathlarger" macro
\newcommand{\BigBox}{\vcenter{\hbox{$\mathlarger{\mathlarger
{\mathlarger{\mathlarger{\mathlarger{\Box}}}}}$}}}
\DeclareMathOperator*{\bigsquare}{\BigBox}
\begin{document}
\[
Z_{\mathcal{T}/G}\equiv \sum_{h\in G}\frac{1}{|G|}
\sum_{g\in G} g {\bigsquare_{\textstyle h}}
\]
\end{document}
Addendum -- Another way to create \BigBox is to load the graphicx package and to run
As you've indicated in a comment that \BigBox isn't an "operator" but a partition function with boundary conditions, I think it makes sense -- from a LaTeX point of view -- to define a macro called, say \partfunc, that takes two arguments, here: g and h, and to align the first argument on the math axis.
Best Answer
I guess the main issues are, how to create a large open square symbol, center it on the math axis, and have it act as a math operator, i.e., place its argument below (and above, if there's an upper limit) the large open square. Sort of like
\sum
, right? A series of\vcenter
,\hbox
, and repeated\mathlarger
instructions applied to the\Box
macro solve these issues. (If you compare the following screenshot with the preceding one, you'll notice that the letterg
to the left of the box is placed ever so slightly higher now.)Addendum -- Another way to create
\BigBox
is to load thegraphicx
package and to runAs you've indicated in a comment that
\BigBox
isn't an "operator" but a partition function with boundary conditions, I think it makes sense -- from a LaTeX point of view -- to define a macro called, say\partfunc
, that takes two arguments, here:g
andh
, and to align the first argument on the math axis.