# how to put equation number at the bottom

i want my eq number 11.72 to appear at the last line.how can i do that?

After a wick rotation we obtain

$$\begin{split} \int \frac{d^4 \mathnormal{k}}{(2\pi)^4}\text{log}(-k^2+m^2)=& i\int \frac{d^4 \mathnormal{k_E}}{(2\pi)^4}\text{log}(k_E^2+m^2)\\ =&-i \frac{\partial}{\partial \alpha}\int \frac{d^4 \mathnormal{k_E}}{(2\pi)^4}\frac{1} {(k_E^2+m^2)^{\alpha}}\Big\vert_{\alpha=0}\\ =&-i \frac{\partial}{\partial \alpha} \Big(\frac{1} {(4\pi)^{d/2}}\frac{\Gamma(\alpha-\frac{d}{2})} {\Gamma(\alpha)}\frac{1}{(m^2)^{\alpha-\frac{d}{2}}} \Big) \Big \vert_{\alpha=0}\\ =& -i\frac{\Gamma(\frac{-d}{2})}{(4\pi)^{d/2}}\frac{1}{(m^2)^{- d/2}} \end{split} \tag{11.72}$$


this is the pdf i obtain.how can i put the eq. number at the bottom?

Like this.

\documentclass{article}
\usepackage{amsmath}
\begin{document}
\begin{align}
\int \frac{d^4 \mathnormal{k}}{(2\pi)^4}\text{log}(-k^2+m^2)=& i\int
\frac{d^4 \mathnormal{k_E}}{(2\pi)^4}\text{log}(k_E^2+m^2)\notag\\
=&-i \frac{\partial}{\partial \alpha}\int \frac{d^4
\mathnormal{k_E}}{(2\pi)^4}\frac{1}
{(k_E^2+m^2)^{\alpha}}\Big\vert_{\alpha=0}\notag\\
=&-i \frac{\partial}{\partial \alpha} \Big(\frac{1}
{(4\pi)^{d/2}}\frac{\Gamma(\alpha-\frac{d}{2})}
{\Gamma(\alpha)}\frac{1}{(m^2)^{\alpha-\frac{d}{2}}} \Big) \Big
\vert_{\alpha=0}\notag\\
=& -i\frac{\Gamma(\frac{-d}{2})}{(4\pi)^{d/2}}\frac{1}{(m^2)^{-
d/2}}
\tag{11.72}
\end{align}
\end{document}


However, I would at least use \log instead of \text{log} and make the differential d distiguishable from the dimension d in dimensional regularization, and same for the imaginary i. Nor would I set the equation number by hand.

\documentclass{article}
\usepackage{amsmath}
\usepackage{cleveref}
\newcommand{\dd}[1]{\mathrm{d}#1}
\numberwithin{equation}{section}
\begin{document}
\section{Dimensional regularization}

\dots

After a Wick rotation we get
\begin{align}
\int\! \frac{\dd^4 \mathnormal{k}}{(2\pi)^4}\log(-k^2+m^2)&= \mathrm{i}\,\int\!
\frac{\dd^4 \mathnormal{k_E}}{(2\pi)^4}\log(k_E^2+m^2)\notag\\
&=-\mathrm{i} \frac{\partial}{\partial \alpha}\int\! \frac{\dd^4
\mathnormal{k_E}}{(2\pi)^4}\frac{1}
{(k_E^2+m^2)^{\alpha}}\Big\vert_{\alpha=0}\notag\\
&=-\mathrm{i}\, \frac{\partial}{\partial \alpha} \Big(\frac{1}
{(4\pi)^{d/2}}\frac{\Gamma(\alpha-\frac{d}{2})}
{\Gamma(\alpha)}\frac{1}{(m^2)^{\alpha-\frac{d}{2}}} \Big) \Big
\vert_{\alpha=0}\notag\\
&= -\mathrm{i}\,\frac{\Gamma(\frac{-d}{2})}{(4\pi)^{d/2}}
\frac{1}{(m^2)^{-
d/2}}
\label{eq:DimRegInt1}
\end{align}

\dots

As shown in \cref{eq:DimRegInt1}, \dots

\end{document}


• thank you sir.i dont understand what u asked? everytime i write \tag{eq.nmbr} in the equation environment.is there a better way?are u sugesting to make a user tag shortcut? Nov 29, 2018 at 4:10
• ok i will use different d's.thanks for the suggestion. Nov 29, 2018 at 4:11
• @jowadulkader Of course there is a better way. If you just remove the \tag from the last line, the equation will have an automatic number, and you could label the equation, e.g. \label{eq:DimRegIntegral1}, and then refer to it with \eqref{eq:DimRegIntegral1} at some other point. If you want the section number in the equation, use \numberwithin{equation}{section}. I changed my example accordingly.
– user121799
Nov 29, 2018 at 4:15

You can simply replace split with aligned and use the [b] optional argument.

I took the liberty to change the size of the big delimiters to \bigg, which look better in this context, in my opinion.

\documentclass{article}
\usepackage{amsmath}

\begin{document}

\begin{aligned}[b] \int \frac{d^4 \mathnormal{k}}{(2\pi)^4}\text{log}(-k^2+m^2)&= i\int \frac{d^4 \mathnormal{k_E}}{(2\pi)^4}\text{log}(k_E^2+m^2)\\ &=-i \frac{\partial}{\partial \alpha}\int \frac{d^4 \mathnormal{k_E}}{(2\pi)^4}\frac{1} {(k_E^2+m^2)^{\alpha}}\bigg\vert_{\alpha=0}\\ &=-i \frac{\partial}{\partial \alpha} \biggl(\frac{1} {(4\pi)^{d/2}}\frac{\Gamma(\alpha-\frac{d}{2})} {\Gamma(\alpha)}\frac{1}{(m^2)^{\alpha-\frac{d}{2}}} \biggr) \bigg \vert_{\alpha=0}\\ & = -i\frac{\Gamma(\frac{-d}{2})}{(4\pi)^{d/2}}\frac{1}{(m^2)^{- d/2}} \end{aligned} \tag{11.72}.

\end{document}


There is an option tbtags to the amsmath package that does this for you:

\documentclass{article}

\usepackage[tbtags]{amsmath}

\begin{document}

$$\begin{split} \int \frac{d^4 k}{(2\pi)^4}\log(-k^2+m^2) &= i\int \frac{d^4 k_E}{(2\pi)^4}\log(k_E^2+m^2)\\ &=-i \frac{\partial}{\partial \alpha}\int \frac{d^4 k_E}{(2\pi)^4}\frac{1} {(k_E^2+m^2)^{\alpha}}\Bigr\vert_{\alpha=0}\\ &=-i \frac{\partial}{\partial \alpha} \Bigl(\frac{1} {(4\pi)^{d/2}}\frac{\Gamma(\alpha-\frac{d}{2})} {\Gamma(\alpha)}\frac{1}{(m^2)^{\alpha-\frac{d}{2}}} \Bigr) \Bigr \vert_{\alpha=0}\\ &= -i\frac{\Gamma(\frac{-d}{2})}{(4\pi)^{d/2}}\frac{1}{(m^2)^{- d/2}} \end{split} \tag{11.72}$$

\end{document}


Other changes

• =& to &= (if you really want to align after = then you need to write ={}& for correct spacing)
• \Big to \Bigl or \Bigr for correct spacing
• \text{log} to \log
• \mathnormal removed, was not used consistently and had no effect

The opposite of the tbtags option is the default centertags. The name tbtags derives from "top or bottom tags"; it will place tags at the top if equation numbers are on the left, or at the bottom if equation numbers are on the right.

The following cleans things up a bit, including using a combination of \nonumber (or \notag) to remove equation numbers where necessary and specify a hard-coded equation number with \tag:

\documentclass{article}

\usepackage{amsmath,mleftright}

\begin{document}

\begin{align}
\int \frac{\mathrm{d}^4 k}{(2 \pi)^4} \log \bigl(-k^2 + m^2 \bigr)
&=  i \int \frac{\mathrm{d}^4 k_E}{(2 \pi)^4} \log \bigl(k_E^2 + m^2 \bigr) \nonumber \\
&= -i \mleft. \frac{\partial}{\partial \alpha} \int \frac{\mathrm{d}^4 k_E}{(2 \pi)^4}
\frac{1}{\bigl(k_E^2 + m^2 \bigr)^\alpha} \mright\rvert_{\alpha = 0} \nonumber      \\
&= -i \mleft. \frac{\partial}{\partial \alpha} \biggl(
\frac{1}{(4 \pi)^{d / 2}} \frac{\Gamma(\alpha - d/2)}{\Gamma(\alpha)}
\frac{1}{(m^2)^{\alpha - d/2}} \biggr) \mright\rvert_{\alpha = 0} \nonumber         \\
&= -i \frac{\Gamma(-d / 2)}{(4 \pi)^{d / 2}} \frac{1}{(m^2)^{-d / 2}} \tag{11.72}
\end{align}

\end{document}


Some pointers:

• It seems unnecessary to use \frac in an exponent/superscript (or subscript for that matter) if it's just single tokens. That is, a/b seems cleaner than \frac{a}{b}.

• Use the (operator) \log instead of \text{log}.

• Stretch extensible delimiters (like \rvert above) using a combination of \mleft...\mright (thanks to mleftright, alternatively use \left...\right, but the spacing might be mildly different/wider).

• align alignment around = should use &=, not =&. You can use the latter, but then you need ={} &.

• @marmot: I saw your answer and it was lacking, so I wrote my own. Why do you think that "I took your answer"?
– Werner
Nov 29, 2018 at 4:25
• @marmot: Yes. And what does that have to do with this?
– Werner
Nov 29, 2018 at 4:29
• @marmot: You've got it completely wrong. The linked post about potrace does not imply that I want credit for something I wrote years before. It provide a link in the right-hand margin of the post thread that identifies linked posts. That way someone who finds the one could potentially be interested in the linked question; roughly a way to identify related posts. Secondly, align is a generic way of aligning equations around some alignment point; there's nothing special about it, nor does your "first use" in some answer void anyone else's use of that in a future answer...
– Werner
Nov 29, 2018 at 4:36
• @marmot: ...the Internet can be a mean place, and there's not need to try and own a space that is free for anyone to contribute. This is not about you, but about the person who asks the question and how contributions add to the community. I did not want to point out all the possible changes that I would implement in your answer, since it wouldn't add value. I just wrote up something that I thought would be more clean and also accurate in terms of the spacing correction originally posted.
– Werner
Nov 29, 2018 at 4:37