\subsection{Configuration Interaction} \label{detci}
                                                                                
Configuration interaction (CI) is one of the most general ways to improve
upon Hartree-Fock theory by adding a description of the correlations
between electron motions.  Simply put, a CI wavefunction is a linear
combination of Slater determinants (or spin-adapted configuration
state functions), with the linear coefficients being determined 
variationally via diagonalization of the Hamiltonian in the given
subspace of determinants.  The simplest standard CI method which improves
upon Hartree-Fock is a CI which adds all singly and doubly substituted
determinants (CISD).  The CISD wavefunction
has fallen out of favor because truncated CI wavefunctions
short of full configuration interaction
are not size-extensive, meaning that their
quality degrades for larger molecules.  MP2 offers a less expensive
alternative whose quality does not degrade for larger molecules and which 
gives similar results to CISD for well-behaved molecules.  CCSD is 
usually a more accurate alternative, at only slightly higher
cost.

For the reasons stated above, the CI code in \PSIthree\ is not optimized
for CISD computations.  Instead, emphasis has been placed on developing
a very efficient program to handle more general CI wavefunctions
which may be helpful in more challenging cases such as highly strained
molecules or bond breaking reactions.  The \PSIdetci\ program is a fast,
determinant-based CI program based upon the string formalism of Handy
\cite{Handy:1980}.  It can solve for
restricted active space configuration interaction (RAS CI) wavefunctions
as described by Olsen, Roos, Jorgensen, and Aa. Jensen \cite{Olsen:1988}.
Excitation-class selected multi-reference
CI wavefunctions, such as second-order CI, can be formulated as RAS CI's.
A RAS CI selects determinants for the model space as those which have no
more than n holes in the lowest set of orbitals (called RAS I) and no
more than m electrons in the highest set of orbitals (called RAS III).
An intermediate set of orbitals, if present (RAS II), has no restrictions
placed upon it.  All determinants satisfying these rules are included
in the CI.  


The \PSIdetci\ program is also very efficient
at full configuration interaction wavefunctions, and is used in this
capacity in the complete-active-space self-consistent-field (CASSCF)
code.  Use of \PSIdetci\ for CASSCF wavefunctions is described
in the following section of this manual.

As just mentioned, the \PSIthree\ program is designed for challenging 
chemical systems for which simple CISD is not suitable.  Because
CI wavefunctions which go beyond CISD (such as RAS CI) are fairly complex,
typically the \PSIdetci\ program will be used in cases where the 
tradeoffs between computational expense and completeness of the 
model space are nontrivial.  Hence, the user is advised to develop
a good working knowledge of multi-reference and RAS CI methods before
attempting to use the program for a production-level project.  This user's
manual will provide only an elementary introduction to the most
important keywords.  Additional information is available in the 
man pages for \PSIdetci.

The division of the molecular orbitals into various subspaces such as
RAS spaces, or frozen vs active orbitals, etc, needs to be clear not
only to the {\tt detci} program, but also at least to the transformation
program (and in the case of MCSCF, to other programs as well).  Thus,
orbital subspace keywords such as {\tt RAS1}, {\tt RAS2}, {\tt RAS3},
{\tt frozen\_docc}, {\tt frozen\_uocc}, {\tt active}, etc., need to be
in the {\tt psi:()} or {\tt default:()} sections of input so they may
also be read by other modules.

\subsubsection{Basic Keywords}
\begin{description}
\item[WFN = string]\mbox{}\\
Acceptable values for determinant-based CI computations in \PSIthree\
are {\tt detci} and, for CASSCF, {\tt detcas}.
\item[REFERENCE = string]\mbox{}\\
Most reference types allowed by \PSIthree\ are allowed by \PSIdetci,
except that {\tt uhf} is not supported.
\item[DERTYPE = string]\mbox{}\\
Only single-point calculations are allowed for {\tt wfn = detci}.
For {\tt wfn = detcas}, first derivatives are also available.
\item[CONVERGENCE = integer]\mbox{}\\
Convergence desired on the CI vector.  Convergence is achieved when the
RMS of the error in the CI vector is less than 10**(-n).  The default is 4
for energies and 7 for gradients.
\item[EX\_LVL = integer]\mbox{}\\
Excitation level for excitations into virtual
orbitals (default 2, i.e. CISD).  In a RAS CI, this is the number
of electrons allowed in RAS III.
\item[VAL\_EX\_LVL = integer]\mbox{}\\
In a RAS CI, this is the additional excitation level for allowing
electrons out of RAS I into RAS II.  The maximum number of holes in RAS I
is therefore {\tt EX\_LVL + VAL\_EX\_LVL}.  Defaults to zero.
\item[FROZEN\_DOCC = (integer array)]\mbox{}\\
Core may be frozen by setting {\tt FREEZE\_CORE}.  To
manually select how many orbitals per irrep to freeze, use the
{\tt FROZEN\_DOCC} keyword.
Should be in {\tt psi:()} or {\tt default:()} sections of input.
The number of lowest energy doubly occupied orbitals in each irreducible
representation from which there will be no excitations.
The Cotton ordering of the irredicible representations is used.
The default is the zero vector.
\item[FROZEN\_UOCC = (integer array)]\mbox{}\\
Should be in {\tt psi:()} or {\tt default:()} sections of input.
The number of highest energy unoccupied orbitals in each irreducible
representation into which there will be no excitations.
The default is the zero vector.
\item[RAS1 = (integer array)]\mbox{}\\
Should be in {\tt psi:()} or {\tt default:()} sections of input.
The number of orbitals for each irrep making up the RAS I space,
from which a maximum of {\tt EX\_LVL + VAL\_EX\_LVL} excitations
are allowed.
This does not include frozen core orbitals.  For a normal
CI truncated at an excitation level such as CISD, CISDT, etc., it is
not necessary to specify this or {\tt RAS2} or {\tt RAS3}.
Note: this keyword must be visible to the \PSItransqt\ program
also so that orbitals are ordered correctly (placing it in 
{\tt default} or {\tt psi} should be adequate).
\item[RAS2 = (integer array)]\mbox{}\\
Should be in {\tt psi:()} or {\tt default:()} sections of input.
As above for {\tt RAS1}, but for the RAS II subspace.  
No restrictions are placed on the occupancy of RAS II orbitals.
Typically this will correspond to the conventional idea of
an ``active space'' in multi-reference CI.
\item[RAS3 = (integer array)]\mbox{}\\
Should be in {\tt psi:()} or {\tt default:()} sections of input.
As above for {\tt RAS3}, but for the RAS III subspace.
A maximum of {\tt EX\_LVL} electrons are allowed in RAS III.
\item[MAXITER = integer]\mbox{}\\
Maximum number of iterations to diagonalize the Hamiltonian.
Defaults to 12.
\item[NUM\_ROOTS = integer]\mbox{}\\
This value gives the number of roots which are to be obtained from
the secular equations.  The default is one.  If more than one root
is required, set {\tt DIAG\_METHOD} to {\tt SEM} (or, for very small cases,
{\tt RSP} or {\tt SEMTEST}).  Note that only roots of the same
irrep as the reference will be computed.  To compute roots of a different
irrep, one can use the {\tt REF\_SYM} keyword (for full CI only).
\item[OPDM = boolean]\mbox{}\\
If {\tt TRUE}, compute the one-particle density matrix for each root.  
By default, it will be written to disk.  Except for 
MCSCF computations (e.g., CASSCF, RASSCF), this will also turn on computation
of dipole moments by default.
\item[TRANSITION\_DENSITY = boolean]\mbox{}\\
If {\tt TRUE}, compute the transition density matrix from the ground
state to each other state obtained in the computation.  By default,
this information will be written to disk.  Transition dipole moments
will be evaluated in {\tt detci}.  Note: only transition densities
between roots of the same symmetry will be evaluated.  {\tt detci} does not
compute states of different irreps within the same computation; to do this,
lower the symmetry using the {\tt subgroup} keyword in {\tt psi:()}
or {\tt default:()} (see section \ref{symm-spec}).
\item[DIPMOM = boolean]\mbox{}\\
If {\tt TRUE}, evaluate the dipole moment for each root (using the expectation
value formula; orbital relaxation contributions are neglected).  This is
an alternative to evaluation using the {\tt oeprop} module, which has
more features.
\item[REF\_SYM = integer]\mbox{}\\
This option allows the user to look for CI vectors of a different irrep
than the reference.  This probably only makes sense for Full CI,
and it is not supported for unit vector guesses.
\item[MPN = boolean]\mbox{}\\
If {\tt TRUE}, compute MPn energies up to nth order, where MAXNVECT = n
controls the maximum order energy computed.  For open-shell systems
(REF = ROHF, WFN = ZAPTN), ZAPTn energies are computed.
\end{description}

For larger computations, additional keywords may be required, as
described in the \PSIdetci\ man pages.

\subsubsection{Arbitrary Order Perturbation Theory}
\PSIthree\ is capable of computing arbitrary order M{\o}ller-Plesset
 perturbation theory (MPn, closed-shell systems) and 
Z-averaged perturbation theory (ZAPTn, open-shell systems) energies,
invoked with MPN = TRUE.  The maximum level of perturbation theory
computed is controlled by MAXNVECT.
Higher order energies (2n - 1 and 2n - 2) can be computed at no additional
computational cost by using WIGNER = TRUE.
By default, the n{th} order energy is saved, but (2n - 1) or (2n - 2) order 
energies can be saved using SAVE\_MPN2 = 1 or SAVE\_MPN2 = 2, respectively.

For open-shell systems, arbitary order ZAPTn energies can be computed using
WFN = ZAPTN and REF = ROHF.
All other options are the same as 
closed-shell MPn.