Search for invisible decays of the Higgs boson produced invector-boson fusion in final states with jets and large missingtransverse energy with the ATLAS detector

The Standard Model of particle physics is a very successful theory, but it leaves some openquestions. Especially the topic of dark matter is a very active field of research and the discoveryof dark matter candidates might be accessible to modern collider experiments. Answering openquestions of the Standard Model is one of the greater goals of this work.The dark matter candidates might interact with the recently discovered Higgs boson andwould appear invisible to a particle detector. This motivates a search for invisible decays ofthe Higgs boson produced in vector-boson fusion. The search is looking for a pair of wellseparated, highly energetic jets and missing transverse energy in the final state. The analysisuses 36.1 fb−1of proton–proton collision data recorded at a centre-of-mass energy of 13 TeV in2015 and 2016 with the ATLAS experiment at the LHC. The main backgrounds are leptonicallydecaying vector bosons. These backgrounds are constrained in dedicated data control regions.The multijet background is small, since it can only result from mismeasurements of thejet transverse momentum, but it is challenging to quantify. The jet response is a measure forthe mismeasurement of jet transverse momenta. To study how well it is simulated in areasof extreme mismeasurements the non-Gaussian tails of these distributions are quantified in acomparison between data and simulation. This is achieved by modelling the Gaussian corewith fits. In order to see the effect in data the momentum balance of jet pairs is considered byusing an extrapolation to pure dijet events. The effort is undertaken with a new jet definition,particle flow jets, as well as topocluster jets. For both of them simulation and data are in goodagreement. This leads to systematic uncertainties small enough to have a negligible impact onthe analysis.The systematic uncertainty resulting from the jet energy resolution is one of the main limitations to the sensitivity of the search. This is addressed with the global sequential calibration(GSC), a simulation-driven method that removes the dependencies of jet momenta on a selection of detector variables in order to improve the jet resolution. The calibration leads to ajet resolution improvement of up to 20%. The GSC is fully derived for particle flow jets forthe first time, allowing performance comparisons between different kinds of jet reconstructionalgorithms.The search is able to derive a new observed (expected) limit on the Higgs to invisible branching fraction of 0.37 (0.28) at 95% confidence level. The results are also interpreted consideringa Higgs portal model, treating the invisible decay products as dark matter candidates. The resulting limits on the cross-section for the DM candidate to interact with an atomic nucleus isbetween 10−46 cm2and 10−42 cm2at 90% confidence level depending on the DM mass andspin