Figure 6.

Pulse sequences used for recording steady-state ¹⁵N-{¹H} nuclear Overhauser effects. In all experiments, a GARP [34] decoupling scheme was initially applied on the ¹⁵N channel for 60 ms followed by a 2 second recovery delay (T_lock) used to lock the magnetic field. Saturated and reference experiments were run in an interleaved manner with the boxed schemes (a) and (b) respectively. (a) A train of 2π/3 pulses was applied on the ¹H channel for 4s with a delay τ_s equal to 5 ms. (b) A recovery delay T is followed by a composite pulse flanked by two water flip back pulses to invert all ¹H polarizations but water ¹H. The total recovery delay is T_lock+T. For comparison with the initial scheme for the reference experiment, bracketed saturation elements were replaced by the delay T. A classic TROSY detection follows for all experiments. Except in the saturation element, all narrow (filled) and (wide) open rectangles represent π/2 and π pulses respectively. Pulse phases are along the x-axis of the rotating frame unless otherwise mentioned. Gaussian shaped pulses are 2 ms and have a Gaussian shape truncated at 1% of the maximal amplitude. A longer duration of 3 ms is recommended if greater selectivity is required. Bell-shaped pulses are sinebell shaped pulses, of 1 ms. The delay τ is 1.792 ms. The phase cycles are φ₁ = {y;−y; −x;x} used with φ₂ = {y} and φ₃ = {x}; and φ₁ = {−y;y; −x;x} used with φ₂ = {−y} and φ₃ = {−x} alternated for every other free induction decay. The phase cycle for acquisition is φ_acq = {x; −x; −y;y}. The duration of acquisition is 120 ms. For the indirect dimension, 75 complex points were recorded with a spectral width of 34 ppm.