2

in written codes 2 equality and 2 pluses are aligned. What I want is to align 2 equality in one hand, and 3 limits on the other hand. I thought that defining another "split" before the second limit would work, but it did not. Would you mind helping me out?

\begin{equation}
\begin{split}
w(z)
&=\sum_{n=0}^{\infty}\frac{(\alpha)_{n}}{n!} z^{n}\int_{\mathcal{C}}t^{(\beta+n)-1}(1-t)^{(\gamma-\beta)-1}dt\\
&=\sum_{n=0}^{\infty}\frac{(\alpha)_{n}}{n!} z^{n}[\lim_{\sigma\rightarrow0}\int_{0}^{1-\sigma}t^{(\beta+n)-1}(1-t)^{(\gamma-\beta)-1}dt\\
&+\underbrace{\lim_{\sigma\rightarrow0}\int_{circle}t'^{(\beta+n)-1}(1-t')^{(\gamma-\beta)-1}dt'}_{0}\\
&+\lim_{\sigma\rightarrow0}\int_{1-\sigma}^{0}t''^{(\beta+n)-1}(1-t'')^{(\gamma-\beta)-1}dt'']
\end{split}
\end{equation}
Stefan Pinnow
  • 29,535
Amir
  • 401
  • 2
  • 8
  • 1
    Welcome to TeX.SE. Please do not post such fragments only. Provide a compilable document –  Jan 25 '18 at 08:04
  • 1
    Split only allows one single alignment point. Switch to alignedat, it will probably be better. It allows more alignment point, at the cost of having to specify the number of alignment points on a line as the argument to the alignedat env (it is similar to the alignat env) – daleif Jan 25 '18 at 08:38
  • Thank you so much for your information. I beg your pardon dr. Hupfer but I could not get your point. Could you be more specific? – Amir Jan 25 '18 at 10:37

1 Answers1

3

Nest an aligned[t] environment within split. Unrelated: the index _{circle} is typed as the product of the six variables c, i, r, c, l, e, with the spacing of variables. Use \text, or \mathrm.

\documentclass[11pt]{article}
\usepackage{mathtools, nccmath}

\begin{document}

 \begin{equation}
\begin{split}
w(z)
&=\sum_{n=0}^{\infty}\frac{(\alpha)_{n}}{n!} z^{n}\int_{\mathcal{C}}t^{(\beta+n)-1}(1-t)^{(\gamma-\beta)-1}dt\\
&=\sum_{n=0}^{\infty}\frac{(\alpha)_{n}}{n!} z^{n}
\biggl[\begin{aligned}[t] & \lim_{\sigma\rightarrow0}\int_{0}^{1-\sigma}t^{(\beta+n)-1}(1-t)^{(\gamma-\beta)-1}dt\\
 & +\underbrace{\lim_{\sigma\rightarrow0}\int_{\text{circle}}t'^{(\beta+n)-1}(1-t')^{(\gamma-\beta)-1}dt'}_{0}\\[-0.5ex]
 & + \lim_{\sigma\rightarrow0}\int_{1-\sigma}^{0}t''^{(\beta+n)-1}(1-t'')^{(\gamma-\beta)-1}dt''\biggr]
\end{aligned}
\end{split}
\end{equation}

\end{document}

enter image description here

Bernard
  • 271,350