Resolution in proton solid state magic angle sample spinning (MAS) NMR is limited by the intrinsically imperfect nature of coherent averaging induced by either MAS or multiple pulse sequence methods. Here, we suggest that instead of optimizing and perfecting a coherent averaging scheme, we could approach the problem by parametrically mapping the error terms due to imperfect averaging in a k-space representation, in such a way that they can be removed in a multidimensional correlation leaving only the desired pure isotropic signal. We illustrate the approach here by determining pure isotropic 1 H spectra from a series of MAS spectra acquired at different spinning rates. For six different organic solids, the approach is shown to produce pure isotropic 1 H spectra that are significantly narrower than the MAS spectrum acquired at the fastest possible rate, with linewidths down to as little as 48 Hz. On average, we observe a 7-fold increase in resolution, and up to a factor of 20, as compared with spectra acquired at 100 kHz MAS. The approach is directly applicable to a range of solids, and we anticipate that the same underlying principle for removing errors introduced here can be applied to other problems in NMR spectroscopy.

The resolution of proton solid-state NMR spectra is usually limited by broadening arising from dipolar interactions between spins. Magic-angle spinning alleviates this broadening by inducing coherent averaging. However, even the highest spinning rates experimentally accessible today are not able to completely remove dipolar interactions. Here, we introduce a deep learning approach to determine pure isotropic proton spectra from a two-dimensional set of magic-angle spinning spectra acquired at different spinning rates. Applying the model to 8 organic solids yields highresolution 1 H solid-state NMR spectra with isotropic linewidths in the 50-400 Hz range.

One key bottleneck of solid‐state NMR spectroscopy is that 1H NMR spectra of organic solids are often very broad due to the presence of a strong network of dipolar couplings. We have recently suggested a new approach to tackle this problem. More specifically, we parametrically mapped errors leading to residual dipolar broadening into a second dimension and removed them in a correlation experiment. In this way pure isotropic proton (PIP) spectra were obtained that contain only isotropic shifts and provide the highest 1H NMR resolution available today in rigid solids. Here, using a deep‐learning method, we extend the PIP approach to a second dimension, and for samples of L‐tyrosine hydrochloride and ampicillin we obtain high resolution 1H‐1H double‐quantum/single‐quantum dipolar correlation and spin‐diffusion spectra with significantly higher resolution than the corresponding spectra at 100 kHz MAS, allowing the identification of previously overlapped isotropic correlation peaks.

The resolution of proton solid-state NMR spectra is usually limited by broadening arising from dipolar interactions between spins. Magic-angle spinning alleviates this broadening by inducing coherent averaging. However, even the highest spinning rates experimentally accessible today are not able to completely remove dipolar interactions. Here, we introduce a deep learning approach to determine pure isotropic proton spectra from a two-dimensional set of magic-angle spinning spectra acquired at different spinning rates. Applying the model to 8 organic solids yields highresolution 1 H solid-state NMR spectra with isotropic linewidths in the 50-400 Hz range.

One key bottleneck of solid‐state NMR spectroscopy is that 1H NMR spectra of organic solids are often very broad due to the presence of a strong network of dipolar couplings. We have recently suggested a new approach to tackle this problem. More specifically, we parametrically mapped errors leading to residual dipolar broadening into a second dimension and removed them in a correlation experiment. In this way pure isotropic proton (PIP) spectra were obtained that contain only isotropic shifts and provide the highest 1H NMR resolution available today in rigid solids. Here, using a deep‐learning method, we extend the PIP approach to a second dimension, and for samples of L‐tyrosine hydrochloride and ampicillin we obtain high resolution 1H‐1H double‐quantum/single‐quantum dipolar correlation and spin‐diffusion spectra with significantly higher resolution than the corresponding spectra at 100 kHz MAS, allowing the identification of previously overlapped isotropic correlation peaks.