Uncut Gem - An Open-Source Hackable Quantum Sensor

Mark Carney, Victoria Kumaran · 2025

Read Uncut Gem - An Open-Source Hackable Quantum Sensor by Mark Carney, Victoria Kumaran (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-18329.txt.

Introduction

Quantum sensors can measure tiny magnetic fields, but most are costly and closed to tinkering. Mark Carney and Victoria Kumaran present Uncut Gem, an open-source sensor anyone can build and modify. Their 2025 paper gives an overview of the full platform. It rests on small flaws in diamond, called NV centers, that react to magnetic fields.

The design uses cheap parts bought off the shelf, arranged in a new hardware layout. Its control software is open source and written with Arduino tools, so it is easy to move and change. The sensor is small and low cost by design. The authors aim it at teaching, lab research, and new quantum technology projects. They frame it as open science, built for a community to improve.

Catalog record

CREATORMark Carney, Victoria Kumaran
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE16 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-18329

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Uncut Gem - An Open-Source Hackable Quantum Sensor Mark Carney, Victoria Kumaran Abstract This work presents an overview of our fully open-source, hackable quantum sensor platform based on nitrogen-vacancy (NV) center diamond magnetometry. This initiative aims to democratize access to quantum sensing by providing a comprehensive, modular, and cost-effective system. The design leverages consumer off-the-shelf (COTS) components in a novel hardware configuration, complemented by open-source firmware written in the Arduino IDE, facilitating portability, ease of customization, and future-proofing the design. By lowering the barriers to entry, our sensor serves as a compact platform for education, research, and innovation in quantum technologies, embodying the ethos of open science and community-driven development. 1 Introduction Quantum technology needs an ‘Apple II’ moment. By this, we refer to the importance that the Apple II holds in the history of the desktop/personal computer. From the first spreadsheet, VisiCalc , to the first graphical adventure game Mystery House , the Apple II was a consumer-grade platform that enabled much of the modern technological world we are now so familiar with [ 16 ] . In a similar vein, the release of the Raspberry Pi was viewed as a key enabling technology in the development of the Internet of Things, driving the decentralization of compute in society even further and giving prominence to the Single Board Computer (SBC) [ 10 ] . Our work has taken inspiration from this concept and applied this reasoning to Quantum Technologies, and sensing in particular. How can we push the boundary of what is conventionally considered to be possible? Open Source software has, in the first quarter of the 21st Century, formed the bedrock of much of our technology stacks and infrastructure. Hoffman et al. [ 8 ] determined that companies would spend around 3.5 \times more on software annually if open-source options did not exist. With this, the importance of open sourcing the key parts of the quantum technology stack become apparent - it drives adoption upwards whilst driving costs down [ 8 ] . This sentiment of increasing openness, interoperability, and institutional momentum is echoed by other open source projects and studies in quantum technology, see [ 2 , 19 ] . Stegemann et al. [ 20 ] managed to get the overall cost of an NV Centre diamond experimentation platform down to under €200, utilising COTS parts and a 3D printer. This made our challenge apparent - could we reduce this cost down even further? And could we fill in some of the gaps and create a full-stack device that is fully open-source, moving away from the optics table and towards a hand-held form factor? We first began designing and developing devices for quantum sensing and magnetometry back in 2023, and this project has been the culmination of 2 years of work towards making at least part of the quantum technology space truly open. 1.1 Brief Overview of NVC Magnetometry The best surveys of the physics behind Nitrogen-Vacancy (NV) centre diamond magnetometry are [ 18 ] for detailed theory, and [ 7 ] for a high-level overview. Figure 1: Energy Level diagram for NV Centre Magnetometry, detailed in section 1.1 . NB - the intersystem crossing (ISC) shelf involving two singlet states with an infrared (IR) transition occurring at 1042 nm, but this is not relevant to our magnetometry. A Nitrogen Vacancy Centre (NV Centre) is the replacement in a diamond lattice of a single carbon (C) atom with a single Nitrogen (N) atom. In lab-grown diamonds this replacement is typically done through electron bombardment [ 13 ] . The mismatch in atomic valencies (4 for C to 3 for N) means that a vacancy is introduced. This defect essentially acts as a virtual atom contained within the lattice structure that has a strong zero-phonon line at 1.945 eV ( \lambda_{ZPL}=637 nm) between the excited states {}^{3}E and ground states {}^{3}G . The hyperfine structures at both these levels yield sub-levels that are split into a lower singlet state m_{s}=0 and two upper states m_{s}=\pm 1 separated by a difference of D=2.87 GHz with no magnetic field present [ 18 ] . The vacancy also has no Doppler shift relative to any interacting photons, which is why its effects may be utilized at room temperature. The ground states Hamiltonian of this, ignoring the hyperfine interactions, is given in [ 20 ] and [ 18 ] as: \mathcal{H}=hDS^{2}_{z}+hE(S^{2}_{y}-S^{2}_{x})+g\mu_{B}\mathbf{B}\cdot\mathbf{S} (1) where z is the parallel axis to the NV centre, D and E are the Zero-Field Splitting parameters, S^{2}_{x} , S^{2}_{y} , and S^{2}_{z} are the spin operators, g is the Landé g -factor, and \mu_{B} is the Bohr magneton. D arises via spin-spin interaction between the two uncoupled electrons in the defect, and E is proportional to the internal strain within the lattice. The resulting resonant frequency \nu (see [ 20 ] ) for the applied microwave field at frequency D is given by \nu=D\pm\sqrt{E^{2}+(\gamma\cdot\mathbf{B}_{||})^{2}} (2) with \gamma=g\mu_{B}/h\approx 28 GHzT -1 [ 18 ] . The important takeaway here is that under this simplified regime, provided E is much less than \mathbf{B}_{||} , the main variable regarding the splitting is the magnetic field \mathbf{B}_{||} and its orientation to the NV centre axis. The Energy-level scheme found in figure 1 summarizes the key physical concepts. We can excite these NV centres with green light \lambda_{E}\approx 532 nm. The NV centres then relax into the upper excited states {}^{3}E , and will become magnetosensitive under the application of microwaves. In practice, we sweep the microwave spectrum surrounding 2.87GHz, utilizing a similar sweep to the sweep in [ 20 ] by starting at 2.614GHz and sweeping up to 3.126GHz in discrete steps of 4MHz. 2 Gap Analysis Many possible design options were reviewed, including the designs and build characteristics from [ 4 , 5 , 9 , 12 , 17 , 20 , 21 ] . The most complete guide was from Stegemann et al. [ 20 ] , which provided a basis for our final design. Upon analysis, however, we found a number of key things were missing from the available designs; The firmware used in NV Center diamond experiments is generally not available. Likewise, a reliable photodetector design schematic is generally not available, with most examples in the literature not always accurate to best electronic design principles. There are few easily configurable diamond mounts, as most require specialist optical mounting apparatus. Lastly, none of the solutions were designed to be ‘hackable’ for rapid deployment or prototyping, with little to no reconfigurable electronics and software. Similarly, none of the designs were geared to being ruggedized towards real world deployment. 3 Design Overview In order to remediate these gaps, we came up with an initial design that made the following key decisions: 1. Use off the shelf components - the only specialist part should be the diamond material and any custom PCBs. 2. Make as many things ‘software problems’ as possible - by careful hardware design we can relinquish most control to a single software source. This massively improves debugging, maintenance, and updating a device. 3. Simplify the design as much as possible. 4. Make the device ‘hackable’ - by this we mean that the device should be fundamentally and easily reconfigurable

Read the complete text