NMR spectrometers are essentially two-way radios. We broadcast radio signals to the sample and detect the radio signal that the sample sends back using a component called the “probe”. NMR probes contain a pair of radio antennas (coils) that pick up the very weak (µV to mV level) signal from the sample. The coils are connected to a tuning circuit made up of wires, capacitors and inductors. The signal from the probe is sent to preamplifiers, which amplify it to mV-V level. There are separate preamplifiers for high frequency (1H and 19F) and low frequency (everything else) nuclei, as well as the lock channel. In traditional or “room-temperature” NMR probes, the electronics and preamplifiers are in separate enclosures and are at ambient temperature. In cryoprobes, the probe electronics and preamplifiers are combined into a single unit and cooled to cryogenic temperatures. Cooling the probe slows the random motion of electrons inside the electronic components, which reduces the electronic noise, or “static”, that they introduce into the NMR signal.
Our cryoprobes operate at 83K using liquid nitrogen as refrigerant. At that temperature, thermal noise from the probe is reduced, which results in a ~300% increase in the signal-to-noise ratio, which is commonly referred to as “sensitivity”.
It’s important to note that only the probe electronics are at cryogenic temperatures. The sample temperature is unaffected during normal operation. On our probes, sample temperature is user controllable between 0 and 130°C (AV500) or -40 and 150°C (AV501) .
Sensitivity and Experiment Time
Due to its low inherent sensitivity, NMR makes extensive use of signal averaging: the spectrometer repeats the same experiment multiple times and sums the resulting data. Because the signal from the sample is the same for each repetition, it simply adds when multiple repetitions are summed. Doubling the number of repetitions doubles the signal level. Noise, on the other hand, is random and therefore it grows in slower: it takes four repetitions to double the noise. The end result is that the signal-to-noise ratio scales with (n/√n) = √n, where n is the number of repetitions. For example, repeating an experiment four times generates four times greater signal, but also doubles the noise, so the signal-to-noise ratio only increases two-fold.
The cryoprobe gets about 2-3 times better signal-to-noise ratio per unit time than room temperature probes and therefore it generates
- 2-3 times better signal to noise ratio in the same amount of time, or
- the same signal to noise ratio in 1/4th to 1/9th the time!
Because of the enhanced sensitivity, you should be able to get publication quality 13C spectra in ~10 minutes on 5 mg of a small molecule. Long 13C experiments should only be used if a 10-minute experiment failed to deliver the desired results.