Chiral logs with staggered fermions
Extracted source text ยท arXiv hep-lat/0209066
\documentclass[twoside]{article}
\usepackage{epsfig,fleqn,espcrc2}
% change this to the following line for use with LaTeX2.09
% \documentstyle[twoside,fleqn,espcrc2]{article}
% if you want to include PostScript figures
\usepackage{graphicx}
% if you have landscape tables
\usepackage[figuresright]{rotating}
% put your own definitions here:
% \newcommand{\cZ}{\cal{Z}}
% \newtheorem{def}{Definition}[section]
% ...
\newcommand{\ttbs}{\char'134}
\newcommand{\AmS}{{\protect\the\textfont2
A\kern-.1667em\lower.5ex\hbox{M}\kern-.125emS}}
% add words to TeX's hyphenation exception list
\hyphenation{author another created financial paper re-commend-ed Post-Script}
\def\refone{Ref.~\cite{CB-SChPT1}}
\def\one{\cite{CB-SChPT1}}
\def\chpt{\mbox{$\chi$PT}}
\def\figref#1{Fig.~\ref{fig:#1}}
\def\Figref#1{Figure~\ref{fig:#1}}
\def\half{{\scriptstyle \raise.2ex\hbox{${1\over2}$}}}
\def\fourth{{\scriptstyle \raise.2ex\hbox{${1\over4}$}}}
\def\eq#1{eq.~(\ref{eq:#1})}
\def\eqs#1#2{eqs.~(\ref{eq:#1}) and (\ref{eq:#2})}
\def\eqsthree#1#2#3{eqs.~(\ref{eq:#1}), (\ref{eq:#2}) and (\ref{eq:#3})}
\def\gtwid{\raise.3ex\hbox{$>$\kern-.75em\lower1ex\hbox{$\sim$}}}
\def\ltwid{\raise.3ex\hbox{$<$\kern-.75em\lower1ex\hbox{$\sim$}}}
\def\ie{{\it i.e.},\ }
\def\eg{{\it e.g.},\ }
\def\et{{\it et al.}}
\def\etc{{\it etc.}}
\def\cA{{\cal A}}
\def\cB{{\cal B}}
\def\cC{{\cal C}}
\def\cD{{\cal D}}
\def\cE{{\cal E}}
\def\cF{{\cal F}}
\def\cG{{\cal G}}
\def\cH{{\cal H}}
\def\cI{{\cal I}}
\def\cJ{{\cal J}}
\def\cK{{\cal K}}
\def\cL{{\cal L}}
\def\cM{{\cal M}}
\def\cN{{\cal N}}
\def\cO{{\cal O}}
\def\cP{{\cal P}}
\def\cQ{{\cal Q}}
\def\cR{{\cal R}}
\def\cS{{\cal S}}
\def\cT{{\cal T}}
\def\cU{{\cal U}}
\def\cV{{\cal V}}
\def\cW{{\cal W}}
\def\cX{{\cal X}}
\def\cY{{\cal Y}}
\def\cZ{{\cal Z}}
\def\prl#1{Phys.\ Rev.\ Lett.\ {\bf #1}}
\def\prd#1{Phys.\ Rev.\ {\bf D#1}}
\def\plb#1{Phys.\ Lett.\ {\bf #1B}}
\def\npb#1{Nucl.\ Phys.\ {\bf B#1}}
\def\jetpl#1{JETP Lett.\ {\bf #1}}
\def\seillac{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 4} (1988)}
\def\fermilab{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 9} (1989)}
\def\capri{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 17} (1990)}
\def\tallahassee{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 20} (1991)}
\def\tsukuba{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 26} (1992)}
\def\amsterdam{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 30} (1993)}
\def\dallas{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 34} (1994)}
\def\bielefeld{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 42} (1995)}
\def\melbourne{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 47} (1996)}
\def\stlouis{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 53} (1997)}
\def\tsukubanew{Nucl. Phys.~B, (Proc. Suppl.), {\bf60A} (1998)}
\def\edinburgh{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 63} (1998)}
\def\boulder{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 73} (1999)}
\def\pisa{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 83-84} (2000)}
\def\bangalore{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 94} (2001)}
\def\berlin{Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.) {\bf 106-107} (2002)}
\def\boston{these proceedings}
%\def\boston{presented at the International Symposium,
%{\it Lattice 2002}, Boston, June 24--29, 2002, to be published
%in Nucl.\ Phys.\ {\bf B} (Proc.\ Suppl.)}
\def\TeV{{\rm Te\!V}}
\def\MeV{{\rm Me\!V}}
\def\GeV{{\rm Ge\!V}}
% declarations for front matter
\title{Chiral logs with staggered fermions\thanks{talk presented by C.\ Bernard at {\it Lattice 2002}; to be published in the proceedings}}
\author{C.~Aubin,\hskip-0.03in
\address{{\vskip-0.10in{\hskip 0.07in Department of Physics, Washington
University, St.~Louis, MO 63130, USA}}} % "a"
C.~Bernard,$\null^{\rm a}$ %a
\hskip-0.01in
C.~DeTar,\hskip-0.03in
\address{Physics Department, University of Utah, Salt Lake City, UT 84112, USA} % "b"
Steven~Gottlieb,\hskip-0.03in
\address{Department of Physics, Indiana University, Bloomington, IN 47405, USA} % "c"
Urs~M.~Heller,\hskip-0.03in
\address{CSIT, Florida State University, Tallahassee, FL 32306-4120, USA} %"d"
K.~Orginos,\hskip-0.03in
\address{RIKEN BNL Research Center, Upton, New York 11973, USA} %""e"
R.~Sugar\hskip0.005in
\address{Department of Physics, University of California, Santa Barbara, CA
93106, USA} %"f"
and
D.~Toussaint\hskip0.005in
\address{Department of Physics, University of Arizona, Tucson, AZ 85721, USA} %""g"
\advance\baselineskip -2pt
} %end \author
\begin{document}
\begin{abstract}
We compute chiral logarithms in the presence
of ``taste'' symmetry breaking of staggered fermions.
The lagrangian
of Lee and Sharpe is generalized and then used to calculate the
logs in $\pi$ and $K$ masses.
%We discuss the calculation of chiral logarithms in the presence
%of the ``taste'' symmetry breaking of staggered fermions.
%We generalize the chiral lagrangian
%of Lee and Sharpe and then compute the
%logs in $\pi$ and $K$ masses.
%Special attention is paid to
%two subtleties: the counting
%of sea quark flavors in loops (which requires the "quark flow
%picture"), and
%the generalization and use
%of the Lee-Sharpe lagrangian for more than one staggered field;
We correct an error in \refone;
the issue turns out to have implications for the
comparison with simulations, even at tree level.
MILC data with three light
dynamical flavors can be well fit by our formulas.
However,
two new chiral parameters, which describe $\cO(a^2)$ hairpin diagrams
for taste-nonsinglet mesons,
enter in the fits.
To obtain precise
results for the physical $\cO(p^4)$ coefficients,
these new parameters
will need to be bounded,
at least roughly.
% (``Gasser-Leutwyler parameters'').
\vspace{-1pc}
\end{abstract}
\maketitle
It has become clear that simulation at rather light quark
mass is crucial for
accurate determination of physical parameters, \eg
heavy-light decay constants \cite{KRONFELD}.
%It is only at light
%mass ($m_\pi/m_\rho \approx 0.3$ to $0.4$, say) that the effects of
%chiral logarithms can be clearly observed and taken in to account.
Since staggered fermions provide the fastest known method of simulating
low-mass
%dynamical
quarks, the systematic
effects associated with this fermion choice
should be studied.
We adopt the following nomenclature:
A single staggered field describes 4 equivalent
``tastes'' of quarks in the continuum limit; taste symmetry is
broken at $\cO(a^2)$.
% with, for example, the 16
%different-taste pseudoscalar mesons breaking into five (approximate)
%multiplets \cite{LEE-SHARPE}.
We use the word ``flavor'' for
different staggered fields; lattice flavor symmetry is exact for equal masses.
MILC's improved staggered (``$a^2$-tad'')
simulations \cite{MILC-RECENT} use three
fields (``$u,d,s$'') and reduce the tastes to one
per flavor by taking $\root 4 \of {\rm Det}$.
We call these ``$2+1$'' flavor simulations since we take
$m_u=m_d\equiv m_\ell$.
MILC data for $m_\pi^2$ {\it vs.}\ quark mass
show clear deviations from linearity, as expected
from chiral logarithms.
However, the detailed behavior at $a\approx0.13$ fm
does not agree with continuum chiral perturbation
theory (\chpt) \cite{GASSER-LEUTWYLER}.
\Figref{nosplit} shows a fit to the continuum forms
for $m_X^2/(m_1+m_2)$, where $X=\pi$ or $K$, and $m_1$, $m_2$
are quark masses.
(We define ``pions'' to have $m_1=m_2$ and ``kaons''
to have $m_2\approx m_s^{\rm phys}$ and $m_1 \not= m_2$.)
The fit is very poor, with a confidence level of $5\times 10^{-5}$.
\begin{figure}[htb]
\null
\vspace{-.6truecm}
%\framebox[55mm]{\rule[-21mm]{0mm}{43mm}}
\includegraphics[bb = 0 0 4096 4096,
%NOTE: to get graphs made with pscax (AT LEAST AS OF 8/24/02)
% must comment out "bounding-box-like" line (e.g "0 0 4096 4096 s")
% immediately after prolog in ps file
%or make it into true bounding box
width=2.9truein]{piK6_nosplit.ps}
\vspace{-1.25truecm}
\caption{\small
$m_{\pi,K}^2/(m_1+m_2)$ {\it vs.}
$m_1 + m_2$, in units of the potential scale $r_1$, for $2+1$ flavor lattices at $a\approx0.13$
fm.
The fit is to the continuum chiral log forms. The upper branch for
$(m_1+m_2)r_1\sim 0.2$ uses the kaon form; the rest
of the curve is the pion form.
}
\label{fig:nosplit}
\vspace{-0.28truein}
\end{figure}
%It is not hard to convince oneself that $\cO(a^2)$ taste-breaking effects at this lattice
%spacing are comparable
%to the deviations seen in \figref{nosplit} and should be included in the
%calculation of the chiral logarithms \one. Fitting to the chiral forms
%that include discretization errors will
%probably be crucial for
%coming generation of much more precise lattice calculations.
%Furthermore, if we are able to show, by comparing simulations to these chiral forms,
%that the staggered theory produces the expected
%chiral behavior in the continuum limit with $\cO(a^2)$
%corrections that are understood and under control, it should
%go a long way toward easing worries about the
%$\root 4 \of {\rm Det}$ trick.
To get good fits, one needs to include the $\cO(a^2)$ taste-breaking effects
into the chiral calculations, \ie we need ``staggered chiral perturbation theory''
(S\chpt).
Such calculations start
with the chiral lagrangian of Lee and Sharpe \cite{LEE-SHARPE}
for 1 staggered flavor (4 tastes).
One defines
\begin{equation}
\Sigma(x) \equiv \exp(i\phi/f) \ ;\quad \phi = \phi_5 \xi_5 + \phi_\mu \xi_\mu + \cdots,
\end{equation}
where $\phi $ is a $4\times4$ matrix of pseudoscalar mesons of various tastes,
and $\xi_5, \xi_\mu, \dots$ are taste matrices.
The Lee-Sharpe lagrangian is then
\begin{equation}
\cL= {f^2\over 8} {\rm tr}(\partial_\mu\Sigma
\partial_\mu \Sigma^\dagger)
-{\mu m f^2\over 4} {\rm tr}(\Sigma + \Sigma^\dagger) + a^2\cV \
\end{equation}
%The first term is the standard $SU(4)_L\times SU(4)_R$ invariant; the second, the mass term.
The taste-breaking potential $\cV$ has various operators with explicit taste matrices:
$ -\cV = \sum_{i=1}^6 C_i O_i$.
For example,
\begin{eqnarray}
O_2 & = & \half[{\rm tr}(\Sigma \Sigma) - {\rm tr} (\xi_5 \Sigma \xi_5 \Sigma) + {\rm h.c.} ]\\
O_5 & = & \half[{\rm tr}(\xi_\nu\Sigma \xi_\nu\Sigma^\dagger) - {\rm tr} (\xi_\nu\xi_5
\Sigma \xi_5 \xi_\nu \Sigma^\dagger)] \ .
\end{eqnarray}
To get NLO chiral results relevant to $2+1$ simulations, we follow a three step
procedure:
{\small
1.\ Generalize the Lee-Sharpe lagrangian to
the 3-flavor case, \ie include
three lattice staggered fields ($u, d, s$), each with
four tastes, and
masses $m_u=m_d\equiv m_\ell \not=m_s$.
%This ``$8+4$'' theory has an $SU(12)_L \times SU(12)_R$ symmetry.
This is an ``$8+4$'' theory.
2.\ Compute the desired quantities (here, $m_\pi^2$ and $m_K^2$) at one loop in the $8+4$ case.
3.\ Adjust the $8+4$ answer ``by hand'' to correspond to the $2+1$
case of interest by
identifying the chiral contributions
that correspond to $n$ virtual quark loops
and dividing them by $4^n$.
}
Step 1 turns out to be non-trivial: \refone\ employed Fierz transformations
to simplify the flavor structure of the generalized potential $\cV$.
However, Ref.~\cite{LEE-SHARPE}
already used the Fierz freedom in deriving $\cV$.
The net result is that the generalized operators $O_2$ and $O_5$ are incorrect in
\refone.
There are actually two operators that correspond
to each of $O_2$ and $O_5$:
\begin{eqnarray}
O_{21} & = & \fourth[{\rm Tr} (\xi_\mu \Sigma ){\rm Tr} (\xi_\mu \Sigma) + {\rm h.c.}]
\nonumber\\
O_{22} & = & \fourth[{\rm Tr} (\xi_\mu\xi_5 \Sigma ){\rm Tr} (\xi_5\xi_\mu \Sigma) +
{\rm h.c.}] \nonumber \\
O_{51} & = & \half{\rm Tr} (\xi_\mu \Sigma ){\rm Tr} (\xi_\mu \Sigma^\dagger) \nonumber \\
O_{52} & = & \half{\rm Tr} (\xi_\mu\xi_5 \Sigma ){\rm Tr} (\xi_5\xi_\mu \Sigma^\dagger) \ .
\label{eq:newops}
\end{eqnarray}
%where ${\rm Tr}$ denotes a $12\times12$ trace.
In the absence of flavor indices, the combinations
$O_{21}+O_{22}$ and $-O_{51}+O_{52}$ can be Fierzed into Lee-Sharpe $O_2$ and
$O_5$, respectively; the other linear combinations turn into other $O_i$.
%However, once flavor is introduced all four linear combinations are independent
%of the remaining operators.
The new operators have a surprising effect: they generate hairpin (disconnected)
diagrams for taste-nonsinglet (but flavor neutral) mesons.
For taste vector and taste axial vector mesons,
this occurs at chiral tree level.
%, as is easy to see by expanding \eq{newops}
%to quadratic order in $\phi$.
The magnitudes of the hairpins are
\begin{equation}
\delta'_V \! \equiv\! {16a^2\over f^2}(C_{21}-C_{51});\; \delta'_A \! \equiv \!
\nonumber {16a^2\over f^2}(C_{22}-C_{52})
\end{equation}
for vector and axial taste, respectively.
%The new (taste-nonsinglet) hairpin operators
%seem to have no reason to vanish at $\cO(a^2)$.
The disconnected hairpin diagrams have not been included in any simulations to date
of taste-nonsinglet mesons.
Thus the Lee-Sharpe lagrangian does not apply to such simulations:
The chiral lagrangian requires more than one flavor to describe particles,
\eg $\pi^+ = u\bar d$, that
have no disconnected contributions.
At 1-loop, we need to iterate three kinds of hairpins on flavor-neutral
internal lines: the standard anomaly hairpin
($m_0^2$) for taste singlets, and the
new hairpins for taste vector and axial vector.
Unlike $m_0^2$, $\delta'_{V,A}$
cannot be taken to infinity.
We therefore have to rediagonalize the mass matrix in the taste $V, A$ channels.
We call, for example, the mass eigenstates in the flavor-neutral,
taste vector channel $\pi^0_V$,
$\eta_V$ and $\eta'_V$.
Techniques in Ref.~\cite{SHARPE-SHORESH} are very useful
for reexpressing the iterated propagator as a sum of simple poles.
Adjustment to go from the $8+4$ to the $2+1$ theory is easy. Every
hairpin interaction on a meson line after the first introduces an additional
virtual quark loop. Therefore, if $F(\Delta)$ is
the sum of hairpin diagrams, just take
$ F(\Delta) \to 4F({\Delta\over4})$ for $\Delta= \delta'_V$, $\delta'_A $ or $\delta$
(where $\delta\equiv m_0^2/(24\pi^2 f^2)$).
Additional diagrams that were
identified as valence diagrams in \refone\ turn out to be
{\it disconnected}
and contribute only to flavor-neutral correlators.
We define $\beta_{K^+_5}$ to be the chiral log term for the Goldstone $K^+$ mass
$(m^{\rm 1-loop}_{K^+_5})^2 /(\mu(m_\ell + m_s))
= 1 + \beta_{K^+_5}/(16\pi^2f^2) + \cdots$,
and similarly for $\beta_{\pi^+_5}$.
%
%\begin{equation}
%{(m^{\rm 1-loop}_{\pi^+_5})^2 \over 2\mu m_\ell }
%& = & 1\! + \!{1 \over 16\pi^2f^2}\Big[\beta_{\pi^+_5}
%+2m_\ell K_3 + (2m_\ell + m_s)K'_4 \Big] \\
%{(m^{\rm 1-loop}_{K^+_5})^2 \over \mu(m_\ell\! + \!m_s)}
%& = & 1 \!+ \!{1 \over 16\pi^2f^2}\Big[\beta_{K^+_5}
%+(m_\ell\!+\!m_s) K_3 + (2m_\ell\! + \!m_s)K'_4 \Big]
%\end{equation}
%(Here $\cdots$ are the analytic terms.)
Our results are then:
\begin{eqnarray}
& \beta_{\pi^+_5} = {2\delta'_V \over m^2_{\eta'_V}-m^2_{\eta_V}}
\Bigg[{m^2_{\eta'_V}-m^2_{S_V} \over m^2_{\eta'_V}-m^2_{\pi_V}}
m^2_{\eta'_V}\ln m^2_{\eta'_V} \nonumber \\
& \quad- {m^2_{\eta_V}-m^2_{S_V} \over m^2_{\eta_V}-m^2_{\pi_V}}
m^2_{\eta_V}\ln m^2_{\eta_V} \Bigg]
-4m_{\pi_V}^2 \ln m_{\pi_V}^2 \nonumber \\
& \quad+ ( V \to A)
+ m_{\pi_I}^2 \ln m_{\pi_I}^2 - {1 \over 3} m_{\eta_I}^2\ln m_{\eta_I}^2 \nonumber \\
%&& \\
& \!\!\!\! \beta_{K^+_5} = 2\delta'_V
\Bigg[{m^2_{\eta'_V}\ln m^2_{\eta'_V}
-m^2_{\eta_V} \ln m^2_{\eta_V} \over m^2_{\eta'_V}-m^2_{\eta_V}}\Bigg] \nonumber \\
&\quad + ( V \to A)
+ {2 \over 3} m_{\eta_I}^2\ln m_{\eta_I}^2\ .
\label{eq:results}
\end{eqnarray}
%$m_{\eta_V}$ and $m_{\eta'_V}$ come the from solution to a quadratic equation,
%and we have used $m_{\pi^0_\mu} = m_{\pi^+_\mu}\equiv m_{\pi_\mu}$.
Here $S$ is the $s\bar s$ meson.
The continuum result is in the taste-singlet channel ($I$); the remainder vanishes in
the limit $\delta'_{V,A}\to0$.
With $\delta'_{V,A}$ as free parameters, these results give excellent fits to
the MILC data.
However, the parameters are
poorly constrained: $\delta'_V$ and $\delta'_A$ wander off in opposite
directions to large values (an order of magnitude larger than
known taste-violating terms in the $\pi^+$ sector).
On the other hand, if we fix $\delta'_V$ or $\delta'_A$ to be of reasonable
magnitude, the fits find reasonable values for the other parameter, too, and the confidence
levels are still excellent.
\Figref{dmu5-.1} shows the fit with $\delta'_A r_1^2$ fixed to $-0.1$;
$\delta'_V r_1^2$ is then found to be $0.25(11)$.
Based on the measured taste violations, the natural size for
these parameters is $\delta' r_1^2\!\sim\!0.2$.
\begin{figure}[htb]
\null
\vspace{-.9truecm}
%\framebox[55mm]{\rule[-21mm]{0mm}{43mm}}
\includegraphics[bb = 0 0 4096 4096,
%NOTE: to get graphs made with pscax (AT LEAST AS OF 8/24/02)
% must comment out "bounding-box-like" line (e.g "0 0 4096 4096 s")
% immediately after prolog in ps file
%or make it into true bounding box
width=2.9truein]{piK6_dmu5-.1.ps}
\vspace{-1.2truecm}
\caption{\small Same as \protect{\figref{nosplit}}, but using the S\chpt\ forms,
\protect{\eq{results}}. The parameter $\delta'_A$ is held fixed to a reasonably
small value in the fit.}
\label{fig:dmu5-.1}
\vspace{-0.30truein}
\end{figure}
We conclude that
S\chpt\ is necessary to fit
current staggered lattice data. The
taste-violating hairpins introduce two new free parameters ($\delta'_{V,A}$)
into the chiral theory; other taste violations, which split
flavor-nonsinglet mesons, are not free parameters in the fits because
they are measured directly in simulations. Unfortunately,
the hairpin parameters are not well constrained, at least at present.
It does however appear to us that even a rough constraint on $\delta'_{V,A}$
(\eg demanding that they be no more than 3 times the known taste-violating terms)
will be sufficient to determine $p^4$ analytic coefficients (Gasser-Leutwyler ``$L_i$'')
with good precision.
We may be able to constrain the hairpin parameters in a variety of ways.
Additional S\chpt\ calculations are underway \cite{AUBIN-CB} that generalize these
results to the quenched and partially quenched cases, and extend them to pseudoscalar decay
constants (of both light-light and heavy-light mesons).
With these calculations in hand,
the direct constraints from all the fits may prove sufficient.
Another approach is to compute the four-quark, taste-violating operators
perturbatively \cite{LEPAGE};
the hairpin parameters may then be estimated by vacuum saturation or lattice calculation
of the matrix elements.
Finally, a direct lattice evaluation of the disconnected hairpin graphs may be possible.
We are grateful to M.\ Golterman, G.\ P.\ Lepage, and S.\ Sharpe for very useful discussions.
Computations were performed at LANL, NERSC, NCSA, ORNL, PSC and SDSC. This work
was supported by the U.S.\ DOE and NSF.
\def\refone{\cite{CB-SChPT1}}
\begin{thebibliography}{9}
\bibitem{CB-SChPT1}
C.\ Bernard,
\prd{65} (2002) 054031.
%hep-lat/0111051.
\bibitem{KRONFELD} A. Kronfeld and S.\ Ryan,
hep-ph/0206058; N.\ Yamada, review talk, \boston.
\bibitem{MILC-RECENT} See, \eg C.~Bernard \et,
%hep-lat/0104002
Phys.\ Rev.\ D {\bf 64}, 054506 (2001).
% \berlin, 412 and 429; and \boston.
\bibitem{GASSER-LEUTWYLER} J.\ Gasser and H.\ Leutwyler,
\npb{250}, 465 (1985).
\bibitem{LEE-SHARPE} W.\ Lee and S.R.\ Sharpe,
%hep-lat/9905023
\prd{60}, 114503 (1999).
\bibitem{SHARPE-SHORESH} S.\ Sharpe and N.\ Shoresh, \prd{62} (2000) 094503.
\bibitem{AUBIN-CB} C.\ Aubin and C.\ Bernard, work in progress.
\bibitem{LEPAGE} G.\ P.\ Lepage, \et, work in progress.
\end{thebibliography}
\end{document}