SEMINARS

NEUTRINO CONFERENCE CYCLE

May 13th at 4:30 PM, Room B
May 20th at 4:30 PM, Room B

Introduction to Neutrino Oscillation from an Experimental Point of View
Speaker: Stefano Dusini
 

Abstract

Neutrino oscillation represents one of the most significant discoveries in particle physics over the past few decades, providing compelling evidence that neutrinos possess non-zero mass and challenging the standard model of particle physics. This seminar presents an experimentalist's perspective on neutrino oscillation phenomena, tracing the historical development from the initial solar neutrino problem to contemporary precision measurements. In the two seminars, we will introduce the theoretical framework of neutrino oscillation, including mixing angles, mass-squared differences, and the PMNS matrix that governs flavour transitions. We will examine the fundamental experimental techniques that have enabled the detection of these elusive particles, including water Cherenkov detectors, liquid scintillator setups, and long-baseline experiments. Special attention will be paid to key experimental milestones such as Super-Kamiokande, SNO, KamLAND, T2K, and NOvA, highlighting how each contributed to our understanding of oscillation parameters. Finally, we will explore future experimental directions, such as DUNE and Hyper-Kamiokande, that promise to further refine our knowledge of neutrino properties.

- - - - - -

June 3rd at 4:30 PM, Room B

How to measure the neutrino mass
Speaker: Riccardo Brugnera

Abstract

Almost a hundred years after the prediction of the neutrinos and about seventy years after its discovery, the neutrino mass is still an unknown parameter of the Standard Model.
In the talk two different techniques used to constrain the neutrino mass will be discussed: the study
of the beta decay end-point and the search for the neutrinoless double beta decay.
The results from the Katrin experiment will be presented as an example of the first method.  
For the neutrinoless double beta the most important experiments in the field will be briefly discussed.
For both techniques the future perspectives will be presented.

- - - - - - -

June 10th at 4:30 PM, Room B

The Sterile Neutrino Puzzle: is LSND still alive?
Speaker: Christian Farnese
 

Abstract

In the last decades different experiments have reported evidences of neutrino oscillations that seems to indicate the possible existence of a fourth state of neutrino, the so called “sterile neutrinos”, associated to large Δm2 of the order of the eV2 and small mixing angles, driving oscillations at short distances. In this seminar the sterile neutrino puzzle will be presented, starting from the results obtained by LSND up to the most recent experiments investigating these anomalies.

________________________________________________________________________________________________________________________________________________________________________________________________

Million-atom simulations of glasses using machine-learned potentials: silicon and phase-change materials

Stephen Elliott of the Physical and Theoretical Chemistry Laboratory
University of Oxford
UK

Monday, January 13th, at 11:00 AM, Room C. Voci

Abstract

The atomic structure of amorphous/glassy materials can only be described in a statistical sense, in terms of correlation functions. In the case of experiment, these are pair distributions, all that can be obtained from single-scattering diffraction. So, modelling the structure, e.g. by molecular-dynamics (MD) simulations, is essential to obtain additional structural information. The most accurate such method is ab initio MD using density-functional theory (DFT), but such models are limited to sizes of only ~1,000 atoms because of cubic size scaling.

Recently, machine-learned (ML) interatomic potentials have been developed which are DFT-accurate and linear-scaling, allowing for the ML-MD simulation of ultra-large structural models over very long time-scales. In this talk, I will describe recent work using ML-MD to construct one-hundred-thousand to one-million-atom models with DFT accuracy, for: i) amorphous silicon (a-Si), where we have shown that the pressure-induced crystallization to the simple-hexagonal phase is via a transient very-high-density-amorphous intermediate phase, and that a-Si contains three distinct types of over-coordinated (five-fold-coordinated) defects; ii) glassy ‘phase-change-memory’ materials (e.g.Ge-Sb-Te), where we have demonstrated the transition from nucleation-limited to growth-limited crystallization with increasing temperature.

____________________________________________________________________________________________________________________________________________________________________________________________________

The computer science physicists need to know
Tim Mattson

Room P1C

November 29 4.30 pm
December 6, 12, 20 4.30 pm

With few exceptions the practice of physics depends on computing.  Learning physics, however, is a full-time job.   Packing your brain with so much math and physics leaves little time to think about (and learn) computer science.

This series of seminars will address this hole in a typical physics education.  We will cover the most vital topics in computer science, topics that every physicist today should know.    The focus will be on core knowledge rather than the practice of engineering good computational solutions.  We will cover computer architecture, parallel computing, two key topics involving computer arithmetic, and the mathematical foundations of data management (both past and hopefully the future).

The lectures will be interactive and driven by the needs of the students.  What we cover and how long we spend on each topic is up to you.   No prior knowledge of computer science is assumed so anyone with a brain full of physics wanting to learn some computer science is welcome to attend.

Planned lectures:

1 Computer Architecture and the need for parallelism

2. High Performance Computing: Parallel systems and the core programming models to program them.

3. Two key facts: Floating Point numbers are not real and your random numbers are almost never random

4. Managing data and the quest for one algebra to rule them all

Biography

Tim Mattson is a parallel programmer obsessed with every variety of science (Ph.D. Chemistry, UCSC, 1985).  He retired in August of 2023 after a 40 year career in scientific computing (the last 30 at Intel).  During his long career, he worked with brilliant people on great projects including: (1) the first TFLOP computer (ASCI Red), (2) MPI, OpenMP and OpenCL, (3) two different research processors (Intel's TFLOP chip and the 48 core SCC), (4) Data management systems (Polystore systems and Array-based storage engines), and (5) the GraphBLAS API for expressing graph algorithms as sparse linear algebra. Tim has over 150 publications including six books on different aspects of parallel computing and is well known as an educator in high performance computing.     

 He is also a recently retired kayak coach and instructor trainer (ACA certified).  His obsession with sea kayaking, including “self-wetting” moments in the ocean, is pretty bad.

COLLOQUIA