Block spins

The work reported here can only be considered as one part of a much larger task. Thus, while keeping as test samples the Beryllium atom and the BeH2 molecule, a similar analysis must be carried out for the aaa 2,-RDM and for all the spin blocks of the A-RDM. 

Similar to spin adaptation each 2-RDM spin block may further be divided upon considering the spatial symmetry of the basis functions. Here we assume that the 2-RDM has already been spin-adapted and consider only the spatial symmetry of the basis function for the 2-RDM. Denoting the irreducible representation of orbital i as T, the 2-RDM matrix elements are given by... 

The OCX spin block of the resulting equations has the following form ... 

The other spin blocks, xf and ff, have a similar stmcture. 

The studies on the spin properties of the 2-RDM and of the second-order correlation matrix [38,49, 50] have shown that for singlet states the ajS block of the 2-RDM completely determines the other two spin blocks of the 2-RDM. In consequence, in these cases, the iterative solution of the 2-CSE may be carried out by working only with the ajS block of the 2-CSE, and the aa and the jS/1 blocks of the 2-RDM are determined in terms of the ajS one. 

As mentioned previously, the off-diagonal elements of the 3-RDM are determined by an algorithm obtained by contracting the 4-RDM. For simplicity sake, the expression given here for the contraction of the 4-RDM (Eq. (71)) corresponds to the spin block, ... 

RDM. The results for each spin block of the 4-RDM are presented in Tables II-V. It should be mentioned that, before carrying the comparison with the FCI 4-RDM, the renormalization procedure previously mentioned was applied. 

As in Section IV.A. the eigenvalues of the 1-RDM must lie in the interval [0,1] with the trace of each block equal to N/2. Similarly, with the a/a- and the jS/jS-blocks of the 2-RDM being equal, only one of these blocks requires purification. The purification of either block is the same as in Section IV.B.2 with the normalization being N N/2 — l)/4. The unitary decomposition ensures that the a/a-block of the 2-RDM contracts to the a-component of the 1-RDM. The purification of Section IV.B.2, however, cannot be directly applied to the a/jS-block of the 2-RDM since the spatial orbitals are not antisymmetric for example, the element with upper indices a, i fi, i is not necessarily zero. One possibility is to apply the purification to the entire 2-RDM. While this procedure ensures that the whole 2-RDM contracts correctly to the 1-RDM, it does not generally produce a 2-RDM whose individual spin blocks contract correctly. Usually the overall 1-RDM is correct only because the a/a-spin block has a contraction error that cancels with the contraction error from the a/ S-spin block. 

Furthermore, properties of the spin components of the 2-G can be obtained by reconsidering the spin properties of the 1-TRDMs. Thus the different spin-blocks of the 1-TRDMs can be related among themselves through the action of the operator S on pure spin states. One therefore has... 

By moving the S operator to the right on the right-hand side (rhs) of Eq. (26), a set of equations hnking the different spin-blocks of the 1-TRDMs is obtained. These equations lead to a set of relations hnking different elements of the spin components of the 2-G matrices. The resulting relations can be classified as follows ... 

As has been mentioned, the MZ purification procedure is based on Coleman s unitary decomposition of an antisymmetric Hermitian second-order matrix described earlier. When applied to singlet states of atoms and molecules, the computational cost of this purification procedure is reduced, since the 2-RDM (and thus the 1-RDM obtained by contraction) presents only two different spin-blocks, the aa- and a/i-blocks (and only one spin-block for the 1-RDM). For the remaining part of this section only this type of state will be treated. 

According to this unitarily invariant decomposition, the different spin-blocks of the trial 2-RDM, which must be corrected, are decomposed as follows ... 

Thus let us consider some particular relations that must be satisfied by the 2-RDM spin-blocks corresponding to a singlet state. It is well known that in this case the aa- and a S-blocks of the 2-RDM are related as follows [100] ... 

This relation imposes severe conditions on the a/i-block of the 2-RDM. Thus this spin-block must satisfy the relations... 

The independence with respect to the type of permutation-symmetry of the decomposition just reported allows one to treat the different spin-blocks of the 2-RDM on an equal footing. Moreover, this decomposition leads to a partitioning of these blocks into three orthogonal parts, which reveal the structure of these blocks with respect to all contraction operations. 

Finally, in order to illustrate the role of the 1-MZ purification procedure in improving the approximated 2-RDMs obtained by application of the independent pair model within the framework of the SRH theory, all the different spin-blocks of these matrices were purified. The energy of both the initial (non-purified) and updated (purified) RDMs was calculated. These energies and those corresponding to a full configuration interaction (full Cl) calculation are reported in Table 111. As can be appreciated from this table, the nonpurified energies of all the test systems lie below the full Cl ones while the purified ones lie above and very close to the full Cl ones. 

In the particular case of singlet states, it can be shown that all the spin-blocks of the 2-CM are proportional to Thus only this spin block is needed to... 

These relations show that each of the spin components presents a one-to-one correspondence with the Ca a spin-block, and therefore with the entire correlation matrix. This is because the 1-RDM—and, consequently, the 1-HRDM— appearing in Eqs. (131) and (132) can be obtained from the different contractions of the spin components. Thus, while it follows from Eqs. (118), (119), (120), (129), and (130) that... 

Figure 6 shows how the 5-representability is attained. Thus it can be seen from this hgure that the ofi Gaa fifi/spin-block converges very satisfactorily on a positive/negative semidehnite matrix. After twenty iterations the lowest/highest eigenvalue of these two matrices is —0.00010 and 0.00022, respectively. As was mentioned in Section II, these conditions are much more exacting than the well-known G-condition. 

RDMs and 2-HRDMs. This may be due to the fact that the different spin-blocks of these matrices are forced to contract correctly and therefore an indirect action on the spin components of the 2-G matrix may occur. On the other hand, the correction of the negativity of the spin components of the 2-G matrix in the AV purification procedure is carried out still more effectively than the negativity correction of the 2-RDM and 2-HRDM. This is illustrated for Li2 and BeH2 molecules in Table VII. The conditions imposed on the spin components of the 2-G and their contractions are essential for the N- and 5-representability of this matrix and, hence, for those of the 2-RDM and 2-HRDM. 

The relation for the M= 1 spin-block follows directly from Eq. (151) by exchanging the spin functions. 

Another extension of this theoretical smdy is the consideration of both an economical and an effective purification strategy for the 4-RDM. The need for such a purification scheme is motivated by the need to have an N- and 5-representable 4-RDM if one wishes to solve the fourth-order modified contracted Schrodinger equation [62, 64, 87]. There have already been several attemps to purify both the 3-RDM and 4-RDM [18, 34, 52]. In particular, a set of inequalities that bound the diagonal and off-diagonal elements of these high-order matrices have been reported [18]. However, the results obtained with this approach within the framework of the fourth-order modified contracted Schrodinger equation (and the second-order contracted Schrodinger equation) were not fully satisfactory because the different spin-blocks of the matrices did not appear to be properly balanced [87, 114]. 

We will consider all the different, although equivalent, forms of decomposition of the 2-RDM (similarly to eq.(lO)), for the different spin-blocks. In this way a set of relations linking the different kinds of correlation terms will be obtained. 

Figure 18. Cooperative spin blocks. When the size N of the particle becomes too big, then thermal activation leads to the formation of cooperative units of size N0.

Despite their similar electronic structures, upconversion processes in Mo + and Re + thus show dramatically different properties that can be related to differences in oscillator strengths due to spin selection rules. The observation of an energetically-resonant but spin-blocked upconversion pathway in Mo + differs from what is normally observed in rare earth upconversion materials. In the latter, spin-orbit coupling is usually so strong that spin selection rules are no longer relevant. 

In organometallic chemistry, many reactions involve several electronic states, in particular, states of different spin. Historically, these reactions were referred as spin forbidden. Later, since reactions with spin state changes were observed to occur, a spin blocking effect that may slow reactions down associated to a hypothetical raise in the activation barriers due to the requirement of spin flip between different state spin multiplicities was proposed.134 However, several kinetic studies of ligand addition reactions with spin changes showed that they can be as fast as spin-allowed processes.135... 

Ensure that a given integral is used in the different spin blocks in the output matrix. 

Figure 5.2. Effective coupling function for a 5-spin block plotted versus the bare coupling parameter g = J/k T.

Figure 5.3. The reduced magnetic susceptibility Xr( > d) of Eq. (5.100) plotted versus the reduced temperature 3 = k TjJ. The continuous curve is a plot of the exact formula [cf Eq. (5.96]. The points indicate results of RG calculations based on the choice of K = 5, that is, on a 5-spin block.

---