Have hierarchical three-body mergers been detected by LIGO/Virgo?

One of the proposed channels of binary black hole mergers involves dynamical interactions of three black holes. In such scenarios, it is possible that all three black holes merge in a so-called hierarchical merger chain, where two of the black holes merge first and then their remnant subsequently merges with the remaining single black hole. Depending on the dynamical environment, it is possible that both mergers will appear within the observable time window. Here we perform a search for such merger pairs in the public available LIGO and Virgo data from the O1/O2 runs. Using a frequentist p-value assignment statistics we do not find any significant merger pair candidates, the most significant being GW170809-GW151012 pair. Assuming no observed candidates in O3/O4, we derive upper limits on merger pairs to be \(\sim11-110\ {\rm year^{-1}Gpc^{-3}}\), corresponding to a rate that relative to the total merger rate is \(\sim 0.1-1.0\). From this we argue that both a detection and a non-detection within the next few years can be used to put useful constraints on some dynamical progenitor models.