mckinsey report
A lot of information here. Measurement is a problem in multiple areas, but can quantum sensing remove/improve the bottleneck better than classical methods?
addressing elephants
…
ronald walsworth
…
intro to nv's
…
qis telecom band emitters silicon
…
the swedish ecosystem
WACQT's stated goals are building a 100-qubit quantum computer and building quantum expertise. who is the commercial customer though? what is the commercial goal? the eu quantum flagship is organizing EUR 1 billion and 5000 researches to work in the 4 main fields of qt. these are defined as physics categories not customer problems. this method consists of physics first, then downstream solutions to customer problems. QSIP does the same, i.e., focused on building an ecosystem, not oriented toward specific problems. same for university holding companies. all of these structures seem to come from the same philosophical idea. if we suppose the physics is the mind and the commerical aspect is the body, then this is clearly a mistake. it becomes aristotle vs plato again. e.g., did elon want to create an ecosystem around space travel or was the goal to get to mars? what do you think works best?
commercial orientation is passive. more revealing questions would be: who would pay, how much, why can't we deploy today, etc. go visit a mining company, geologists, medical imaging centers. the customer context is deferred and this rationalized as appropriate caution given the technology's maturity. also, the ecosystem seems focused on "prestige," i.e., citation counts, conferences, grants, which all are second hand values. this is seen by statements such as "sweden should be a leading nation in XXX," but where leading refers to nr of publications. if you recall the distinguishing characteristic of mr. big in the "prime mover" article, it is the living for the appearance of success. the risk of the ecosystem is more of inward focus. a report on the quantum ecosystem in sweden mentioned three startup founders that requested anonymity, why?
what would quantum sensing look like at a deployable scale in a specific market? analogized to edison, he would envision power stations, wiring, cost competitiveness with gas, in other words, he would think about the entire electrical grid. what is the equivalent for the qs in the swedish ecosystem? see the whole, not only the parts.
high goals produce high performance. 100-qubit computer is a high physics goal, sure. and how do you measure the success and failure of that goal? likely qubit output and publication count. the job of a leader is defining what these terms mean, properly they ...
n2n-models. quantum instrument companies are selling to quantum researchers. most quantum research is funded by WACQT, which is funded by wallenberg foundation and participating companies. the quantum instrument companies are also funded by WACQT.
on independence. is the goal of WACQT a result of independent thinking or is it mimicking ibm and google? it seems highly unlikely that sweden can win a global quantum computing race.
i think the missing element is a specific person. everything in the ecosystem is described collectively and top-down. let's invert the relationship. start from specific customer => specific problem => difference between what classical sensing provides and what they need => get the minimimum quantum sensing performance that erases the difference => get the deployment constraints of the sensor => do some physics & engineering that meet the expectations. besides low noise factory ab, spectracure seem to have been aristotelian in their approach by solving. whatever answer you get from these questions will be far more valuable than "achieve the highest possible sensitivity."
quantum illumination by seth lloyd
…
gaussian-state quantum illumination saikat guha et al
…
two-mode squeezing radar by loung
…
microwave quantum illumination by shabir
…
integrated and portable magnetometer based on nv
building a portable fiber-integrated nv-diamond magnetometer achieving 344 pT/√Hz sensitivity. the materials used:
| Subsystem | What they used |
|---|---|
| Diamond | 0.8 × 0.8 × 0.5 mm³, 99.97% carbon-12 enriched HPHT diamond |
| NV creation | 2 MeV electron irradiation, annealed at 1000 °C for 2 h |
| NV density | about 0.4 ppm NV⁻ |
| P1 impurity | about 2.0 ppm |
| Pump laser | 521.9 nm fiber-coupled compact laser, 23.5 mW at fiber end |
| Optics | single-mode fiber + GRIN rod lens |
| Microwave structure | double split-ring resonator on Rogers RO3010 PCB |
| MW tuning | resonance tunable around 2.8–3.0 GHz |
| Detection | two photodiodes with DBR optical filters |
| Electronics | logarithmic transimpedance amplifier + RedPitaya demodulation |
| Bias field | tapered Helmholtz coils, about 1.07 mT at 0.5 A |
this is run by hitting the nv spin resonance with the microwave, flourescence changes, demodulate the flourescence signal and use the zero crossing as the magnetic field readout. by readout we mean anyhting that turns the nv diamond's response into a signal on the computer. some useful parameters:
| Parameter | Their optimum |
|---|---|
| Modulation depth | 40 kHz |
| Microwave power | 58 µW |
| Integration time | 20 ms |
| Modulation frequency | 1 kHz |
| Best shielded sensitivity | 344 pT/√Hz |
| Best unshielded estimate | about 424 pT/√Hz |
in the paper they calculate a photon detection rate of 26 pT/√Hz but the measured one is 344 pT/√Hz. the noise is therefore 13 times worse in actual experiment. they mention several causes for this. but what is the main constraint? well assuming that the goal is to build a portable integrated NV magnetometer with better magnetic sensitivity, then anything that limits the system from achieving that goal is a constraint. i think it is the readout floor since without any magnetic field the instrument output jitters around 364 pT/√Hz. note how in the paper many things are optimized ... but we need to know which instrument causes the noise so check the laser, photodiode, mw frequenecy, temp diff, etc. redesign the experiment so we can see the bottleneck clearly.
note to self: redesign their experiment to see the bottleneck clearly.