Speaker
Description
We employ a tree-tensor-network strong-disorder renormalization group (tSDRG) to study the disordered antiferromagnetic spin-$3/2$ Heisenberg chain with bond-alternations. Extended SDRG predicts two distinct random-singlet regimes in the weak- and strong-disorder limits, characterized by different effective-spin structures ($S_{\mathrm{eff}}=1/2$ and $S_{\mathrm{eff}}=3/2$), separated by an intervening multicritical point $P_4$.[Ref:1,2] Here we locate $P_4$ using the distribution of spin correlations and the theoretical prediction of the critical exponent is verified via finite-size scaling. With bond alternation, we further map the phase boundaries between distinct VBS phases; notably, these boundaries converge toward the $P_4$ location. In the weak-disorder regime, however, our tSDRG results deviate from theoretical expectations, which we attribute to the reduced scale separation that undermines the strong-disorder assumption and renders the local decimation approximation less reliable. Overall, our results validate the predicted multicriticality and clarify the practical regime of applicability of tSDRG, motivating refinements or complementary approaches in the weak-disorder limit.