CCL:G: Which is the best way to work with N-OXIDES?

["Miguel A. Quiliano" <rifaximina===gmail.com>]

 Sent to CCL by: "Miguel A. Quiliano" [rifaximina a gmail.com]
 Hello CCL users.
 I am a new user of this forum so please be patient jeje.
 Recently, I have a problem, but basically it`s a doubt. I have performed
 geometrical optimization of one compound (by the way I`m not an expert using
 QM). It`s a derivate N-oxide.
 In the paper is:                 after Gaussian, using Gaussview I see this:
                o                                            o
       / - \   //                                   ---\    //
       I     C-C                                  /      C- C
       I Ar  I  \                                I       I   \
       I     I   C- R                            I  Ar   I    C - R
       I     I  //                               I       I   //
       \     C- N(+)                              \      C - N
        \ - /  /                                   ---- /   //
              O(-)                                          O
 I set a simple "# opt pm3 geom=connectivity". I want to study REDOX
 properties, but N(+)-O(-) seems totally different to N=O, Although you can see
 that nitrogen continuous with 5 bonds, I missed the N-O
 What do you think about this?? is correct the set of bonds between N=O? I know
 that partial double bond is present but is it acceptable for my objetive? If you
 were me, what should I do to prevent a posible mistake? What would you have
 done, if you have wanted a good  conformation to work??
 Sorry for the long message, perhaps this problem is very easy for you.
 I would be grateful if someone can help me or give me advices
 Thanks in advance.
 M.Q