I've noticed the same thing with the sparsediag program. However, if you put in the number of states (MAXSTATES=50 for instance) and sweeps (NUMSWEEPS=4 or some such), and use dmrg you will get the correlations. If you need more accuracy just increase the number of states (and sweeps if necessary).

________________________________________ From: comp-phys-alps-users-bounces@phys.ethz.ch [comp-phys-alps-users-bounces@phys.ethz.ch] On Behalf Of Cristian Degli Esposti Boschi [cristian.degliesposti@unibo.it] Sent: Wednesday, May 13, 2009 8:02 AM To: comp-phys-alps-users@phys.ethz.ch Subject: [ALPS-users] Local measurements and correlations with sparsediag

Dear all,

I would like to study one-site and two-sites correlation functions in certain variants of the 1D Hubbard model.

I have tried to compute the local densities (total, up and down) in a small Hubbard chain (10 sites) with open boundary conditions, so that the one-point observables are non-trivial. However, all the local observables return the value 0. The same happens if I try to compute the correlation cdag_up:c_up (or down). The input file is as follows:

MODEL="fermion Hubbard"; LATTICE="open chain lattice"; LATTICE_LIBRARY="./lattices.xml"; MODEL_LIBRARY="./models.xml"; CONSERVED_QUANTUMNUMBERS="Nup; Ndown"; MEASURE_LOCAL[nup]=n_up MEASURE_LOCAL[ndown]=n_down MEASURE_LOCAL[ntot]=n MEASURE_CORRELATIONS[c+up_cup]="cdag_up:c_up" MEASURE_CORRELATIONS[c+dn_cdn]="cdag_down:c_down" U=1; {L=10;Nup_total=3;Ndown_total=3}

Is there a reason why the correlations are not properly computed and/or a mistake in the input file?

Thanks for your time. Cristian -- ___________________________________________________ Cristian Degli Esposti Boschi CNR, CNISM, Unita' di Ricerca di Bologna, c/o Dipartimento di Fisica, Universita' di Bologna viale Berti-Pichat, 6/2, 40127, Bologna, Italia tel. ++39 051 2095114 fax ++39 051 2095113 e-mail: cristian.degliesposti -AT- unibo.it web: http://www.df.unibo.it/fismat/theory ___________________________________________________