Skip to content

wonmor/Atomizer-Swift-Challenge

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

icon-2

Atomizer AR

This public archive was not intended for use in production. Please read the license carefully before using any parts or segments of this source code.

Developed by John Seong, 2023. My WWDC23 Swift Student Challenge Submission. Built using Xcode 14 on macOS Ventura, designed to run on iPhone, iPad, and ARM-based Macs.

iOS, iPadOS, and macOS

346091630_962156048139650_5797215412402448016_n


Atomizer is a state-of-the-art SwiftUI app that visualizes atomic and molecular orbitals. It features atomic orbitals in the form of electron density, and molecular orbitals in which you can view it in AR.

  • ARKit and SceneKit are used to render the 3D models. Object occlusion is also supported on LiDAR-enabled devices.

  • Using Apple's powerful ML-trained Vision framework, I was able to achieve object spawning by hand detection, in which the molecule will come right into your hand like Thor's hammer, all in Augmented Reality.

  • For the electron coordinate data, it uses the ElectronVisualized API I created from scratch by myself, using Python and Flask.


Watch the Demo: https://www.youtube.com/watch?v=kHcdvyaqslU

GitHub repo of the ElectronVisualized API that I created: https://github.com/wonmor/ElectronVisualized

  • I used ASE and GPAW to get electron density data, using Density Functional Theory (DFT). For the molecular orbitals, I used PySCF to get the molecular orbitals, using Hartree–Fock (HF) theory.

  • For atomic orbitals, I used the spherical harmonics to compute the radial part of the atomic orbitals. Then, I sampled the wavefunction by using the Metropolis-Hastings algorithm.