DEAP-3600 paper explores additional constraints on dark matter search

In physics research, theory and experiments advance knowledge in a cyclical pattern: theorists develop theories to explain observed phenomena, and experimentalists in turn develop experiments to test whether these theories are accurate, which in turn generate new data for theorists to explain. For the most part though, theorists and experimentalists work separately, as much as they are informed by each other’s work.  

Recently, experimentalists working on DEAP-3600 partnered with a theorist to see how data from the Gaia sky map could be used to set some additional limits on the dark matter search DEAP is performing. The results of their work were published in Physics Review D.

There are certain assumptions that all dark matter collaborations make when analyzing their data – this is necessary to be able to compare results from one experiment to others in order to look for confirmation of results and effectively ‘rule out’ certain search areas. 

First, most dark matter experiments assume that dark matter follows a ‘typical distribution’ throughout our galaxy. Since the Big Bang, dark matter has had time to settle out and now it isn’t doing anything dramatic: it’s found its place in the galaxy and has a stable orbit. 

Second, dark matter experiments assume that when a dark matter particle collides with a nucleus in a detector the collision isn’t dependent on how fast the particles are moving.   

The typical distribution assumption is especially important for dark matter searches because it informs the way different experiments ‘exclude’ certain search areas (these search areas equate to dark matter particle energy and size). 

The elastic-collision assumption means that within the data, scientists expect to see a dark matter signal within a particular energy range and mainly look there – similar to how you look for a certain star in the part of the sky where you know it will be. 

Each time an experiment looks in this expected area of its data and doesn’t find dark matter, another mass/size is ruled out for potential dark matter particles, setting further constraints on what dark matter could look like.  

When theorists took a look at the data collected from DEAP-3600 for this paper, they decided to question these assumptions and propose a different set of constraints the experimental data sets on dark matter particles. According to Shawn Westerdale, a physicist at Princeton University and one of the lead collaborators on the paper, “The work for this paper started when the three of us were discussing how we might be able to place constraints on other, less-common, particle physics models using the latest null results reported by DEAP-3600.” 

The Gaia map has created the biggest star catalogue ever with 1.7 billion stars mapped. This image below is not a photo but a compilation of every star Gaia has mapped, with each one representing one pixel. 

When you catalogue over a billion stars, you start to notice when stars in a certain area act a bit wonky. Enter the Gaia Sausage. This is an area of nearby stars where the distribution of stars’ velocities suggests that there was a galactic collision at some time in the past. As the name suggests, it also looks something like a sausage. This galactic collision means that the dark matter clouds in each of those galaxies also collided, leaving the dark matter in that area with an atypical distribution. Being relatively nearby, the sausage is a good candidate for where dark matter that could be detected by experiments like DEAP-3600 could come from. However, since there is good evidence that dark matter in this region is atypically distributed, typical exclusion plots relying on assumption 1 might miss it.  

Dark matter has not yet been directly detected, and one of the only things we know for sure about it is that it has mass and doesn’t interact with the electromagnetic spectrum. These uncertainties mean that dark matter particles might not collide with other particles (argon atoms in the case of DEAP) in the expected fashion. If, for example, there is some yet-undiscovered force at play between dark matter and standard matter, then the velocity and momentum of those particles might determine whether they collide. Theorists explored what these particle interactions could look like in the data. It affected both how sensitive the detector could be, and where in the experiment’s data the signal would appear, meaning that analysis done with assumption 2 could again miss a dark matter interaction.  

Westerdale said of what they found, “As we started exploring this, we saw that in some cases, very realistic variations of the astrophysical model can actually dramatically impact the way we interpret dark matter search results. In other words, we found that the impact of variations in the astrophysical and particle physics models were not independent of each other, and that both components are needed to really understand experimental results.” The success of this collaboration and the importance of combining theoretical and experimental perspectives to really understand everything the data is saying indicate that moving forward with the search for dark matter, theorists and experimentalist can and should work together more often. 

Ariel Zuñiga-Reyes, a PhD student from UNAM in Mexico, has been working with the DEAP-3600 experiment since 2018 and was a main participant in the analysis. “This research examined how current DEAP-3600 limits are modified due to the presence of substructures in the solar neighborhood,” explained Zuñiga-Reyes. Examinations like this add context to the data DEAP generates, and can help researchers more accurately analyze their data for dark matter signals.   

This wasn’t the first time that theorists have teamed up with a dark matter experiment to explore the data further. As Aaron Vincent, a physicist at Queen’s University who worked on the project said, “theorists are good at recycling and getting everything out of the data.” One of the outcomes of this paper was, according to Vincent, “nicely parameterized types of [potential] dark matter, structured in a way that isn’t overwhelming and could be used by other experiments in the future”. 

Read the paper on the arXiv preprint server or in the Physical Review D journal.