% our theorists, on DM-e scattering in Si: https://arxiv.org/pdf/1509.01598.pdf
% DOE BRN: https://science.osti.gov/-/media/hep/pdf/Reports/Dark_Matter_New_Initiatives_rpt.pdf
% PDG DM: https://pdg.lbl.gov/2020/reviews/rpp2020-rev-dark-matter.pdf
% last SENSEI: https://arxiv.org/pdf/2004.11378.pdf


%\documentclass[presentation]{beamer}
\documentclass[presentation,aspectratio=169]{beamer}
% Replacing 'presentation' with 'handout' in the above line
% will produce 4 slides per page.
% The hyperref option makes it possible to include hyperlinks.

\mode<presentation> {
    \usetheme{Boadilla}
    \setbeamercovered{transparent}
}

\usepackage[english]{babel}
\usepackage[latin1]{inputenc}
\usepackage{times}
\usepackage[T1]{fontenc}
%\usepackage{wrapfig}
\usepackage{amsmath}

%\usepackage{movie15}
%\usepackage[autoplay]{animate}


\mode<handout>{
    \usepackage{pgfpages}
    \pgfpagesuselayout{4 on 1}[letterpaper,landscape,border shrink=5mm]
\setbeamercolor{background canvas}{bg=black!10} }

\setbeamertemplate{footline}
{
    \leavevmode%
    \hbox{%
        \begin{beamercolorbox}[wd=.333333\paperwidth,ht=2.25ex,dp=1ex,center]{author
            in head/foot}%
            \usebeamerfont{author in head/ foot}\insertshortauthor%&\approx& (\insertshorti
        \end{beamercolorbox}%
        \begin{beamercolorbox}[wd=.333333\paperwidth,ht=2.25ex,dp=1ex,center]{title
            in head/foot}%
            \usebeamerfont{title in head/foot}\insertshorttitle
        \end{beamercolorbox}%
        \begin{beamercolorbox}[wd=.333333\paperwidth,ht=2.25ex,dp=1ex,right]{date in
            head/foot}%
            \usebeamerfont{date in head/foot}\insertshortdate\hspace*{2em}
            \insertframenumber / \inserttotalframenumber\hspace*{2ex}
    \end{beamercolorbox}}%
    \vskip0pt%
}

%gets rid of bottom navigation bars
%\setbeamertemplate{footline}[frame number]{}

%gets rid of bottom navigation symbols
%\setbeamertemplate{navigation symbols}{}

%gets rid of footer
%will override 'frame number' instruction above
%comment out to revert to previous/default definitions
%\setbeamertemplate{footline}{}

\newcommand{\backupbegin}{
    \newcounter{framenumberappendix}
    \setcounter{framenumberappendix}{\value{framenumber}}
}
\newcommand{\backupend}{
    \addtocounter{framenumberappendix}{-\value{framenumber}}
    \addtocounter{framenumber}{\value{framenumberappendix}} 
}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\usepackage{hyperref}
\hypersetup{
    colorlinks=false,
    urlcolor=blue,
    }
    %linkcolor=black,

\title[SENSEI]{SENSEI}

\author[Sho Uemura]{Sho Uemura}
\institute{Tel Aviv University\\for the SENSEI Collaboration\\\vspace{1cm}{\tiny SU was supported in part by the Zuckerman STEM Leadership Program}}
\date[June 16, 2021]

%\titlegraphic{
%\includegraphics[height=0.1\textheight]{SLAC_Logo}\hspace*{4.75cm}~
%\includegraphics[height=0.1\textheight]{partner_logo_v2}
%}

\begin{document}

%\setcounter{framenumber}{2} 
\begin{frame}
    \titlepage
\end{frame}

%Detector concept (target mass, material, shape, setup, sensor, temperature, threshold, resolution, surroundings, exposure history, ...)

%Observed energy spectrum in d.r.u = counts/(keV*kg*day) to make the figures easily comparable; total rate (without normalizing it to mass or exposure)

%Time dependency and known correlations of the excess

%Are there approaches to explain the excess?

%What are other known backgrounds?

%Are there uncertainty estimates for the energy scale?




%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{frame}{Intro to SENSEI}
    \begin{itemize}
        \item Detector very similar to DAMIC (fully depleted CCD, collecting holes from e-h pairs produced by ionization), but with non-destructive ``Skipper'' readout
            \begin{itemize}
                \item Sub-electron charge resolution: self-calibrating, discriminates 0, 1, 2, \dots$e^-$
            \end{itemize}
        \item We are sharing our most recent published result
            \begin{itemize}
                \item Test run: one production CCD, one month, shallow underground site with partial shield
                \item \href{https://arxiv.org/abs/2004.11378}{arXiv:2004.11378}, published in PRL
            \end{itemize}
        \item We treat 1, 2, 3, 4 $e^-$ events as separate channels with different cuts and analyses
    \end{itemize}
    \begin{center}
        \includegraphics[width=0.6\textwidth]{ccd}
        \includegraphics[width=0.35\textwidth]{peaks-crop}
    \end{center}
\end{frame}

\begin{frame}{The SENSEI CCDs}
    \begin{columns}
        \column{0.5\textwidth}
        \begin{itemize}
            \item High-resistivity silicon $675$ $\mu$m thick (no backside thinning), active area $1.59\times 9.42$ cm$^2$ (1.925 grams)
            \item $6144\times 886$ pixels (split in 4 quadrants), 15 $\mu$m pitch
            \item Largely identical to DAMIC
                \begin{itemize}
                    \item Same material, processing, pixel geometry
                    \item Different overall dimensions, readout
                \end{itemize}
                %\item Designed by LBNL MSL, fabricated by Teledyne DALSA
            \item Huge improvements (in dark current and amplifier luminescence) over previous SENSEI CCDs fabricated parasitically on lower-quality silicon
        \end{itemize}
        \column{0.5\textwidth}
        \includegraphics[width=\textwidth]{ccds}
    \end{columns}
\end{frame}

\begin{frame}{CCD package}
    \begin{columns}
        \column{0.45\textwidth}
        \begin{itemize}
            \item Front side of CCD is glued (Epotek 301-2) and wirebonded to silicon pitch adapter, CCD touches nothing else
            \item Pitch adapter serves multiple functions:
                \begin{itemize}
                    \item Electrical interface to copper-Kapton flex cable
                    \item Mechanical support
                    \item Thermal connection to copper tray
                \end{itemize}
            \item PA (with silicon spacers) is held in copper tray: loose fit at all points except for copper leaf spring
            \item PA and spacers are made from the same high-resistivity silicon as the CCDs
        \end{itemize}
        %\begin{center}
        %\includegraphics[width=\textwidth]{emcal_location}
        %\end{center}
        \column{0.55\textwidth}
        \includegraphics[width=\textwidth]{sensei_module}

        \includegraphics[width=0.48\textwidth]{schemside}
        \includegraphics[width=0.48\textwidth]{schemtop}
    \end{columns}
\end{frame}

\begin{frame}{Setup in MINOS}
    \begin{columns}
        \column{0.47\textwidth}
        \begin{itemize}
            \item Shielding design adapted from DAMIC: cylindrical vacuum vessel with lead ``plug'' above the CCD
            \item CCD at 135 K, biased at 70 V
            \item The copper tray is not light-tight and may admit blackbody photons
                \begin{itemize}
                    \item Most surfaces the tray sees are close to room T
                \end{itemize}
        \end{itemize}
        \column{0.25\textwidth}
        %\includegraphics[width=\textwidth]{minos_lead}
        \includegraphics[width=\textwidth]{MINOS_inside}
        \column{0.28\textwidth}
        %\includegraphics[width=\textwidth]{minos_lead}
        \includegraphics[width=\textwidth]{MINOS_setup}
    \end{columns}
\end{frame}

\begin{frame}{Location and shielding}
    \begin{itemize}
        \item Shallow underground site (MINOS cavern at Fermilab), $\sim105$ m ($\sim225$ mwe)
        \item Lead shield thickness: 3 inches above, 3-4 inches to sides (but with gaps), 0 below
            \begin{itemize}
                \item Lead is of good quality (leftover DAMIC shielding) but not ancient
                \item Nothing outside the lead, very little inside (few mm of copper)
            \end{itemize}
    \end{itemize}
    \begin{center}
        \includegraphics[width=0.6\textwidth]{minos_location}
        \includegraphics[width=0.3\textwidth]{minos_lead}
    \end{center}
\end{frame}

\begin{frame}{Data taking}
    \begin{columns}
        \column{0.5\textwidth}
        \begin{itemize}
            \item 22 images in the blinded dataset (Feb. 25 --- Mar. 20, 2020)
                \begin{enumerate}
                    \item Erase/e-purge: manipulate bias voltages to eliminate latent image and suppress dark current
                    \item 20 hours exposure
                    \item 5.15 hours readout
                        \begin{itemize}
                            \item During readout, the image continues to accumulate hits and charge
                            \item Skipper noise scales as $1/\sqrt{N}$: at $N=300$ (13 ms/pixel), noise of $\sim 0.14e^-$
                        \end{itemize}
                \end{enumerate}
            \item Shown: 1/6th of one quadrant
                \begin{itemize}
                    \item Two quadrants (19.93 g-day) acceptable for 1, 2 $e^-$ analyses; add part of a third (27.82 g-day) for 3, 4 $e^-$
                \end{itemize}
        \end{itemize}
        \column{0.5\textwidth}
        \includegraphics[width=\textwidth]{ds9_skp_72000secs_exp_run10_NSAMP_300_36_hdu2_annotated}
    \end{columns}
\end{frame}

\begin{frame}{The data}
    \begin{columns}[T]
        \column{0.5\textwidth}
        \begin{itemize}
            \item Single-electron: background-dominated
            \item Multiple electrons (2$e^-$ single-pixel, 3/4$e^-$ cluster): low-background
                \begin{itemize}
                    \item Coincidence of $1e^-$ processes
                    \item True multi-$e^-$ processes
                \end{itemize}
        \end{itemize}
        \includegraphics[width=\textwidth]{spectrum}
        \column{0.5\textwidth}
        \begin{itemize}
            \item Our measured charge spectra are continuous, but the underlying charge distribution is quantized, so our EXCESS dataset is binned by charge
                \begin{itemize}
                    \item The Gaussian widths come from readout noise, not physics
                    \item We count 1311.7 1$e^-$ events because we correct for 0/1$e^-$ misID
                \end{itemize}
            \item We see five 2$e^-$ pixels, and no 3/4$e^-$ clusters
            \item Geometric efficiency: some multi-$e^-$ events will be lost when they diffuse
                \begin{itemize}
                    \item 22.8\% for 2$e^-$ to stay in one pixel, 76.1\%, 77.8\% for 3, 4$e^-$ to form a contiguous cluster
                \end{itemize}
        \end{itemize}
    \end{columns}
\end{frame}

\begin{frame}{Known 1$e^-$ backgrounds (paper on arXiv soon)}
    \begin{columns}[T]
        \column{0.5\textwidth}
        \begin{itemize}
            \item Well-controlled backgrounds:
                \begin{itemize}
                    \item Bleeding (incomplete charge transfer): masked
                    \item CCD defects (hot pixels, bleed columns): identified and masked
                    \item Amplifier light: not significant
                    \item Intrinsic dark current (generated during exposure by thermal excitation): extrapolated, not significant at 135K
                    \item Spurious charge (generated during readout): measure and subtract from 1$e^-$ rate
                \end{itemize}
        \end{itemize}
        %\includegraphics[width=\textwidth]{spectrum}
        \column{0.5\textwidth}
        \begin{itemize}
            \item Backgrounds that may remain:
                \begin{itemize}
                    \item Halo (excess charge near high-energy events): masked, but maybe not completely
                    \item Loose clusters (localized charge excesses not associated with HE events): masked for 2+$e^-$ analyses but not 1$e^-$
                    \item ``Extrinsic dark current'' (other exposure-dependent charge): shielding-dependent
                \end{itemize}
        \end{itemize}
    \end{columns}
\end{frame}

\begin{frame}{Spurious charge}
    \begin{columns}
        \column{0.55\textwidth}
        \begin{itemize}
            \item Measurements with shorter exposures show a limiting value for the CCD charge: accounts for $649 \pm 47.5$ counts in the 1$e^-$ data (about half of the total)
                \begin{itemize}
                    \item Since this contribution is independent of exposure time, it is not physics
                    \item Consistent with ``spurious charge'' --- charge excited by voltage transitions during readout
                \end{itemize}
            \item For the SENSEI result we subtracted the $2\sigma$ LL on $649 \pm 47.5$ to get 758 counts
        \end{itemize}
        \column{0.4\textwidth}
        \includegraphics[width=\textwidth]{SC}

        \includegraphics[width=\textwidth]{sc_vs_v}
    \end{columns}
\end{frame}

\begin{frame}{Localized charge from high-energy events}
        \begin{itemize}
                %\item Bleeding: charge ``left behind'' by charge transfer
            \item Halo: excess of charge near high-charge pixels
                \begin{itemize}
                    \item Probably near-bandgap photons from Cherenkov radiation and electron-hole recombination (see \href{https://arxiv.org/abs/2011.13939}{arXiV:2011.13939})
                \end{itemize}
            \item Loose clusters: regions with high charge density
                \begin{itemize}
                    \item May be Cherenkov photons (reflected, or generated outside of CCD)
                \end{itemize}
                %\item We apply a tight cut (>60 pixels from any high-charge pixel) for the 1$e^-$ analysis, looser (>20 pixels) for the 2/3/4$e^-$ analyses
                %\item After cuts, charge density of $3.188(90)\times 10^{-4}$ $e^-$/pixel
        \end{itemize}
        \begin{center}
        \includegraphics[width=0.4\textwidth]{Halo}
        \includegraphics[width=0.3\textwidth]{photondistance}
        \end{center}
        %\column{0.2\textwidth}
        %\includegraphics[width=\textwidth]{bleed}

        %\includegraphics[width=\textwidth]{ds9_skp_72000secs_exp_run10_NSAMP_300_36_hdu2}
\end{frame}

\begin{frame}{1$e^-$ rate vs. shielding}
    \begin{columns}
        \column{0.75\textwidth}
        \begin{itemize}
            \item We have data with and without the outer ring of lead bricks
            \item Factor of 3 reduction in the rate of high-energy tracks $\to$ similar reduction in the 1$e^-$ rate
                \begin{itemize}
                    \item Radiation generates charge in halos, in loose clusters, and pseudo-uniformly
                    \item Again, Cherenkov+recombination produces a rate estimate consistent with what we observe
                        %\item Better shielding will very likely further reduce our 1$e^-$ rate
                \end{itemize}
        \end{itemize}
        \begin{center}
            \includegraphics[width=0.45\textwidth]{Background_wineNcheese}
            \includegraphics[width=0.48\textwidth]{DCvsBackground}
        \end{center}
        \column{0.25\textwidth}
        \includegraphics[width=\textwidth]{minos_lead}
    \end{columns}
\end{frame}

\begin{frame}{Multi-$e^-$ backgrounds}
    \begin{columns}
        \column{0.5\textwidth}
        \begin{itemize}
            \item Low-energy tails from high-energy events (these do not contribute significantly in this dataset):
                \begin{itemize}
                    \item Compton: A gamma ray can create arbitrarily small-energy electron recoils
                    \item Partial charge collection: Our CCDs have a highly-doped backside layer where much of the charge is lost to recombination
                \end{itemize}
            \item Photons over the 2$e^-$ energy threshold (Cherenkov or luminescence from the epoxy?)
        \end{itemize}
        \column{0.4\textwidth}
        \includegraphics[width=\textwidth]{compton}

        \includegraphics[width=\textwidth]{pcc}
    \end{columns}
\end{frame}

\begin{frame}{Counts and rates}
        \begin{itemize}
            \item 1$e^-$: we subtracted spurious charge, the rest is consistent with environmental effects (``extrinsic DC'')
            \item 2$e^-$: we did not subtract coincidences
            \item Exposure is corrected for cuts and geometric efficiency
            \item Energy scale: we conservatively assumed $E_e \to (1 + Floor[(E_e - 1.2 \mathrm{eV})/(3.8 \mathrm{eV})]) e^-$
                \begin{itemize}
                    \item This is electron recoil energy
                \end{itemize}
        \end{itemize}
        %    \includegraphics[width=0.9\textwidth]{diffusion}
        %\includegraphics[width=\textwidth]{spectrum}

        \begin{center}
            {\small
            \begin{tabular}{l|r|r|r|r}
                & counts & known bkgd & exposure [g-day] & rate [/g-day]  \\
                \hline
                1$e^-$ & 1311.7 & $649 \pm 47.5$ & 1.38 & 450  \\
                2$e^-$ & 5 & --- & 2.09 & 2.39  \\
                3$e^-$ & 0 & --- & 9.03 & 0  \\
                4$e^-$ & 0 & --- & 9.10 & 0  \\
                %\hline
            \end{tabular}
            }
        \end{center}
        %1$e^-$: 1311.7 counts, minus the $2\sigma$ LL on $649 \pm 47.5$ 

        %\begin{center}
        %    {\small
        %    \begin{tabular}{l|r}
        %        & 90\% CL  \\
        %        \hline
        %        1$e^-$ & 525.2 events/g-day  \\
        %        2$e^-$ & 4.449 events/g-day  \\
        %        3$e^-$ & 0.255 events/g-day  \\
        %        4$e^-$ & 0.253 events/g-day  \\
        %        %\hline
        %    \end{tabular}
        %    }
        %\end{center}
        %\includegraphics[width=\textwidth,angle=270]{simulations_in_4umPixelCCD}
        %\includegraphics[width=\textwidth,page=1]{SENSEI_size_calibration}

        %\includegraphics[width=0.4\textwidth,page=12]{selected_muons}
        %\includegraphics[width=0.55\textwidth]{diffusion_sim15}

\end{frame}


%%%%
% PCC
% Compton
% other applications
%%%%

\appendix
\backupbegin

\begin{frame}{Backup: Intrinsic dark current}
    \begin{columns}
        \column{0.55\textwidth}
        \begin{itemize}
            \item Subtracting the exposure-independent charge, our 1e- rate is $1.59(16)\times 10^{-4}$ $e^-$/pixel/day
            \item Intrinsic dark current is the usual suspect
                \begin{itemize}
                    \item Thermal generation of electron-hole pairs, mediated by lattice defects 
                \end{itemize}
            \item However:
                \begin{itemize}
                    \item Extrapolation from higher temperatures (dashed black line) predicts $\ll 1\times 10^{-5}$ $e^-$/pixel/day at our operating temperature of 135 K
                        \begin{itemize}
                            \item Suppressing surface dark current gets us from red data points to blue
                        \end{itemize}
                \end{itemize}
            \item High-quality silicon has made this a subdominant background
        \end{itemize}
        \column{0.45\textwidth}
        %\includegraphics[width=\textwidth]{DC_vs_T}
        \includegraphics[width=\textwidth]{DC_vs_T_subtracted}
    \end{columns}
\end{frame}


\backupend

\end{document}