\documentclass[12pt]{article}
\usepackage{graphicx}
\usepackage{url}
\usepackage[left]{lineno}
\linenumbers

\title{SENSEI gets quiet}

\author{Sho Uemura}

\date{}

\begin{document}
\maketitle
What makes for a good dark matter detector?
It has a lot in common with a good teleconference setup: you need a sensitive microphone and a quiet room.
In a recently released result, the SENSEI experiment demonstrated both the high sensitivity and the low background needed for an effective search for low-mass dark matter.

\begin{figure}[!ht]
    \begin{center}
        \includegraphics[width=0.55\textwidth]{sensei_module} 
        \includegraphics[width=0.4\textwidth]{inner-000} 
    \end{center} 
    \caption{
        Left: the new SENSEI Skipper-CCD module.
        Right: the CCD module in the detector vessel.
    }
    \label{fig:module}
\end{figure} 

The SENSEI detector is a 5.4-megapixel CCD (charge-coupled device) made of 2 grams of silicon, which we are operating underground in the MINOS cavern at Fermilab.
If a dark matter particle collides with one of the electrons in the silicon, the energy transferred to the electron will be converted to electron-hole pairs.
This charge is the signal we are looking for; the smaller a signal SENSEI can detect, the broader the range of dark matter models it can test.

To observe small dark matter signals, the first thing we need is a sensitive detector.
In other words, we must be able to detect a small signal and consistently distinguish it from an empty detector.
As demonstrated in previous work, the LBNL-designed Skipper-CCDs used in SENSEI can count exactly how many electrons are in each pixel.

\begin{figure}[!ht]
    \begin{center}
        \includegraphics[width=\textwidth]{gaussianidad} 
    \end{center} 
    \caption{
        The SENSEI Skipper-CCDs are capable of almost arbitrarily high charge resolution, and therefore perfect sensitivity to charge.
        In this test data, taken with a very long acquisition time, the readout noise is 0.04 elementary charges RMS and is totally Gaussian.
    }
    \label{fig:peaks}
\end{figure} 

Second, we need low background - the rate of signal-like events from causes other than dark matter has to be small.
A sensitive detector with high background is like a studio microphone in a noisy room.
Even if the microphone can pick up a whisper, you might need to shout in order to be heard over the washing machine!
It's important to demonstrate that a detector can achieve low background rates before you scale up to a larger experiment with the same technology, because otherwise you are just going to scale up your background rate.
This is where SENSEI has made progress.

Previous dark matter searches by SENSEI used prototype CCDs, which had high sensitivity but also high backgrounds.
These CCDs were fabricated parasitically as part of a production run for astronomical CCDs, so they were not made with the highest quality silicon.
As a result, there was significant background from a process called dark current, where impurities in the silicon randomly create electron-hole pairs.

Our new dark matter search is the first result from our new ``science-grade'' CCDs, which were fabricated in a dedicated production run for SENSEI with high-quality silicon.
We also reduced the amount of radiation that hits the CCD by adding extra shielding around the experiment.
The result was a decrease in event rates compared to the previous search with a prototype CCD.
The rate of one-electron events decreased from $3.3\times 10^4$ to $4.5\times 10^2$~events/gram-day, and we see fewer two-electron events (5, down from 21) in a much larger exposure (2.09 gram-days, up from 0.043).
We also see no three- or four-electron events - just as in the previous search, but with a larger exposure.

\begin{figure}[!ht]
    \begin{center}
        \includegraphics[width=0.45\textwidth]{SENSEI_MINOS2020_FDM1-Paper-final}
        \includegraphics[width=0.45\textwidth]{SENSEI_MINOS2020_FDMq2-Paper-final}
    \end{center}
    \caption{
        New SENSEI limits on the DM-electron scattering cross-section, $\overline{\sigma}_e$,
        as a function of DM mass, $m_\chi$, for two different models of the DM-electron interaction (left and right).
    }
    \label{fig:dm}
\end{figure}

\begin{figure}[!ht]
    \begin{center}
        \includegraphics[width=\textwidth]{spectrum} 
    \end{center} 
    \caption{
        The new SENSEI data for one- and two-electron events.
    }
    \label{fig:spectrum}
\end{figure} 

This result is important in two ways.
First, the rates that we have measured are record lows for a silicon detector.
They therefore set the world's strongest limits on dark matter interactions with electrons, across a wide range of models.
Second, this is encouraging news for the full-scale SENSEI experiment that we are building at SNOLAB, deep underground.
Our science-grade CCDs work as well as we could have hoped, and we expect our background rates to be even lower at SNOLAB.
We look forward to sharing more great science from SENSEI in the near future!

Learn more from our preprint (\url{https://arxiv.org/abs/2004.11378})
or our presentation at the Wine and Cheese Seminar
(\url{https://theory.fnal.gov/events/event/new-results-from-sensei-zoom/}).
\end{document}