$DCCORR group       (optional)
        relevant for MPLEVL=2
        relevant for CCTYP=LCCD, CCD, CCSD, CCSD(T), R-CC
 
    This group controls the linear-scaling DC-based MP2 or
CC calculations. In this method, subsystem correlation
energy is evaluated in each subsystem by means of subsystem
MOs.  Total correlation energy is obtained by summing up
subsystem contributions.
 
    The present implementation allows only RHF reference.
DC-MP2 calculations can be run in parallel (using CODE=DDI,
IMS, or SERIAL in $MP2), but DC-CC is limited to serial
execution.  DC-MP2 with CODE=IMS is only compatible with
DIRSCF=.TRUE.  Coupled cluster code is only available for
CCTYP=LCCD, CCD, CCSD, CCSD(T), or R-CC.  No solvation
models are available.  This group must be given if the
"double-hybrid" DFT is used (e.g., DFTTYP=B2PLYP).
 
    Note: Although $DANDC input is usually used together to
select subsystem and buffer information, DC-SCF calculation
is not indispensable to perform DC correlation calculation.
You can perform DC correlation calculation without DC-SCF
by setting DCFLG=.FALSE. in $DANDC and DODCCR=.TRUE.
 
    For more information (and references), see $DANDC.
 
 
DODCCR = a flag to activate DC-MP2/CC calculation. This is
         forced to be .TRUE. if DCFLG=.TRUE. in $DANDC.
         This keyword enables to perform DC-MP2/CC
         calculation after standard (non-DC) RHF.
 
RBUFCR = buffer radius used in DC-MP2/CC calculation. This
         value should be in the unit given by UNITS option
         in $CONTRL.  By default, RBUFCR is set to be equal
         to BUFRAD in $DANDC.
         This keyword is mainly used to perform so-called
         dual-buffer DC-MP2/CC calculations, see the paper
         on the DC-CC method for more details.
 
RMKORB = a flag to remake orbitals in each subsystems. This
         is forced to be .TRUE. if RBUFCR is different from
         BUFRAD in $DANDC or standard HF calculation was
         performed. Apart from these cases, RMKORB=.FALSE.
         by default. This keyword is meant for debug
         purposes.
 
HFFRM  = a flag to use the Fermi level determined in the
         preceding HF calculations even when RMKORB=.TRUE.
         (default=.FALSE.)
         The Fermi level is used to classify the subsystem
         orbitals into occupied and virtual ones. Usually,
         this option does not change the results except
         for the use of diffuse basis functions.
 
WOCC   = a parameter determining proportion of occupied
         contribution.  This should be between 0 and 1.
         The proportion of virtual contribution becomes
         [1 - WOCC]. (default=1.0)
         This is forced to be 1.0 in DC-CC calculation,
         except when WOCC=0.0, which only calculates
         virtual contribution.
         We recommend 1.0 to obtain accurate results.
 
ONLYOC =         a flag to disable MP2 calculation for
                 virtual contributions. This is forced to
                 be .FALSE. if WOCC is not 1.0, and to be
                 .TRUE. in DC-CC calculation.
       = .TRUE.  Performs DC-MP2 calculation only for
                 occupied contributions. This option will
                 accelerate the CPU time. (default)
       = .FALSE. Performs DC-MP2 calculation for occupied
                 and virtual contributions.
 
ITPART =    specifies the partitioning for (T) correction.
            This is only relevant to CCTYP=CCSD(T) or R-CC.
       = XY (two-digit integer)
            uses [X,Y] type partitioning defined in the
            following article: J.Chem.Phys.131,114108(2009)
            (default=00)
 
ISTCOR =   restart option for DC-MP2/CC.
       = 0 does DC-MP2/CC calculation from the beginning
           (default).
       = n reads subsystem correlation energies
           corresponding to subsystem 1-(n-1) from input
           and perform DC-MP2/CC calculation from n-th
           subsystem.  $MP2RES and $CCRES inputs are
           required for DC-MP2 and DC-CC calculations,
           respectively.
 
FZCORE =   a flag to freeze core electrons in DC-MP2 or
           DC-CC calculation. Other frozen orbital options
           options such as NACORE in $MP2 and NCORE in
           $CCINP do not pertain to DC-MP2/CC calculations.
           The default is .TRUE. to freeze cores.