newtxmath and breqn problem: Viewing capital Greek math incorrectly

Question

I am writing a document with too many long equations and I adopted the use of breqn package for its simplicity in writing such long equations when alignment is not a big restriction. The problem is when I use newtxmath I can not get correct Greek math capitals. If I use Computer Modern, the problem disappears.

Here is my MWE which shows the problem:

\documentclass[a4paper,oneside,12pt]{book}
\usepackage{lmodern}
\usepackage{newtxtext} 
\usepackage[scaled=0.92]{helvet}
\usepackage{textcomp} 
\usepackage{amsmath}
\usepackage[cmintegrals,bigdelims]{newtxmath}
%\usepackage{flexisym}
%\usepackage{breqn}
\usepackage{bm} % load after all math to give access to bold math

\begin{document}
Some alternative text goes here.
%
\begin{equation}\label{eqdsb}
\varphi^{}_{\mathrm{DSB-SC}}(t)\ne\Delta\varphi^{}_R\ne\Delta\theta^{}_R
\end{equation}

%\begin{dmath}
%I_{3} = 
%%
%\sum_{k=1}^{\infty}{\frac{\left(\frac{c}{\bar{\gamma}\pi ^{\frac{3}{2}}}\right)(-1)^{k+1}\left (\frac{K}{\sqrt{a_{10}}}\right )^{2k-1}\Gamma \left (p+\frac{1}{2}\right )}{(2k-1)(k-1)!\Gamma (p+1)}}\left \{E_{4k-2}\left (\frac{1}{\bar{\gamma }}\right )\delta (p)+\frac{1}{2\bar{\gamma}\sqrt{\pi}} \right . \times  
%%
%\sum_{p=1}^{\infty}\left(\frac{(-1)^{p}(2k-1)(2k-3)\ldots (2k-(2p-3))}{2^{p}p!} \right . \times  
%% 
%\left.\left.  \frac {b^{p}\Gamma(p+\frac{1}{2})}{\Gamma(p+1)} E_{4k-2}\left (\frac{1}{\bar{\gamma}}\right )\right )\right \}. 
%\end{dmath}

\begin{equation}
\begin{split}
I_{3} &= \\
%
&\sum_{k=1}^{\infty}{\frac{\left(\frac{c}{\bar{\gamma}\pi ^{\frac{3}{2}}}\right)(-1)^{k+1}\left (\frac{K}{\sqrt{a_{10}}}\right )^{2k-1}\Gamma \left (p+\frac{1}{2}\right )}{(2k-1)(k-1)!\Gamma (p+1)}}\left \{E_{4k-2}\left (\frac{1}{\bar{\gamma }}\right )\delta (p)+\frac{1}{2\bar{\gamma}\sqrt{\pi}} \right . \times\\  
%
&\sum_{p=1}^{\infty}\left(\frac{(-1)^{p}(2k-1)(2k-3)\ldots (2k-(2p-3))}{2^{p}p!} \right . \times \\ 
% 
&\left.\left.  \frac {b^{p}\Gamma(p+\frac{1}{2})}{\Gamma(p+1)} E_{4k-2}\left (\frac{1}{\bar{\gamma}}\right )\right )\right \}.
\end{split}
\end{equation}

\end{document}

I commented out the problematic parts. The desired output should be like this:

If I uncomment the commented lines, I get Gamma as \prime symbol. I appreciate your help.

Answer 1

The flexisym package makes some assumptions about the font slots characters and symbols are taken from. However, newtxmath has different ideas from the standard setup and for good reasons, because it provides many more symbols.

For fixing uppercase Greek letters here is a workaround:

\documentclass[a4paper,oneside,12pt]{book}
\usepackage{lmodern}
\usepackage{newtxtext} 
\usepackage[scaled=0.92]{helvet}
\usepackage{textcomp} 
\usepackage{amsmath}
\usepackage[cmintegrals,bigdelims]{newtxmath}
%\usepackage{flexisym}
\usepackage{breqn}
\usepackage{bm} % load after all math to give access to bold math

\makeatletter
% newtxmath uses mathgroup lettersA for uppercase Greek
\edef\mg@Greek{\hexnumber@\symlettersA}
\makeatother

\begin{document}

Some alternative text goes here.
%
\begin{equation}\label{eqdsb}
\varphi^{}_{\mathrm{DSB-SC}}(t)\ne\Delta\varphi^{}_R\ne\Delta\theta^{}_R
\end{equation}

\begin{dmath}
I_{3} = 
%
\sum_{k=1}^{\infty}{\frac{\left(\frac{c}{\bar{\gamma}\pi ^{\frac{3}{2}}}\right)(-1)^{k+1}\left (\frac{K}{\sqrt{a_{10}}}\right )^{2k-1}\Gamma \left (p+\frac{1}{2}\right )}{(2k-1)(k-1)!\Gamma (p+1)}}\left \{E_{4k-2}\left (\frac{1}{\bar{\gamma }}\right )\delta (p)+\frac{1}{2\bar{\gamma}\sqrt{\pi}} \right . \times  
%
\sum_{p=1}^{\infty}\left(\frac{(-1)^{p}(2k-1)(2k-3)\ldots (2k-(2p-3))}{2^{p}p!} \right . \times  
% 
\left.\left.  \frac {b^{p}\Gamma(p+\frac{1}{2})}{\Gamma(p+1)} E_{4k-2}\left (\frac{1}{\bar{\gamma}}\right )\right )\right \}. 
\end{dmath}

\begin{equation}
\begin{split}
I_{3} &= \\
%
&\sum_{k=1}^{\infty}{\frac{\left(\frac{c}{\bar{\gamma}\pi ^{\frac{3}{2}}}\right)(-1)^{k+1}\left (\frac{K}{\sqrt{a_{10}}}\right )^{2k-1}\Gamma \left (p+\frac{1}{2}\right )}{(2k-1)(k-1)!\Gamma (p+1)}}\left \{E_{4k-2}\left (\frac{1}{\bar{\gamma }}\right )\delta (p)+\frac{1}{2\bar{\gamma}\sqrt{\pi}} \right . \times\\  
%
&\sum_{p=1}^{\infty}\left(\frac{(-1)^{p}(2k-1)(2k-3)\ldots (2k-(2p-3))}{2^{p}p!} \right . \times \\ 
% 
&\left.\left.  \frac {b^{p}\Gamma(p+\frac{1}{2})}{\Gamma(p+1)} E_{4k-2}\left (\frac{1}{\bar{\gamma}}\right )\right )\right \}.
\end{split}
\end{equation}

\end{document}

Usage of breqn may simplify the input of complex formulas, but doesn't improve their quality. In my opinion, the output of breqn is in no way better than the manual one. Bot formulas are, of course, horrible.

In particular, breaking after \frac{1}{2\bar{\gamma}\sqrt{\pi}} is wrong in both cases. You can fix it in the manually broken version; fixing something with breqn is almost impossible.