# How to (correctly) manage utf-8 in bibliographic references

users,

I have a persistent problem when I try to construct bibliographic references using 'exotic' accents. An example is provided below, with the error message. I am aware that this is a known problem, but I don't understand what the core of the problem is (I just suppose that it has something to do with multibyte segmentation) and, anyway, I just found some hints at rather cryptic workarounds.
Anticipated thanks for enlightening me.

\documentclass[11pt,a4paper]{article}

\usepackage[T1]{fontenc}
\usepackage[utf8]{inputenc}

\usepackage{amsmath}
\usepackage{amsfonts}
\usepackage{amssymb}

\usepackage[left=2.50cm, right=2.50cm, top=2.50cm, bottom=2.50cm]{geometry}

\usepackage[authoryear,round]{natbib}

\begin{document}

On this notion see the work of Búbür (2009).  \\\
On this notion see \cite{Búbür2009}.

\begin{thebibliography}{widestlabel}

\bibitem[\protect\citeauthoryear{Búbür}{Búbür}{2009}]{Búbür2009}
\newblock Búbür, Bub (2009).
\newblock A theory of everything plus something else.
\newblock \textit{Journal of Unfocusing} 12, pp.\ 22-39.

\end{thebibliography}

\end{document}


pdflatex loudly complains as soon as it sees the first accented u:

! Missing \endcsname inserted.
\IeC
l.21 ...year{Búbür}{Búbür}{2009}]{Búbür2009}

The control sequence marked <to be read again> should
not appear between \csname and \endcsname.

! Improper alphabetic constant.
\IeC
l.21 ...year{Búbür}{Búbür}{2009}]{Búbür2009}


A one-character control sequence belongs after a  mark. So I'm essentially inserting \0 here.

• Welcome to TeX.SX! You can have a look at our starter guide to familiarize yourself further with our format. – Heiko Oberdiek Apr 2 '14 at 10:24
• The error is in using accents in the key, which is just an arbitrary label; use \cite{bubur2009} and change the key in the bib file. – egreg Apr 2 '14 at 10:30

BibTeX it­self is an ASCII-only pro­gram; there is, how­ever, a ver­sion that copes with 8-bit char­ac­ter sets. How­ever, BibTeX’s fa­cil­i­ties rapidly run out as one moves away from sim­ple ASCII (for ex­am­ple, in the var­i­ous na­tional sort­ing rules for lan­guages ex­pressed in dif­fer­ent parts of ISO-8859 — the “ISO Latin” se­ries). For more flex­i­bil­ity, the user is urged to con­sider us­ing biber with bibla­tex to type­set its out­put. In par­tic­u­lar, it is best to avoid BibTeX in favour of bibla­tex`, if at all pos­si­ble.