Geminal coupling constant of formaldehyde protons in NMR

I read in a book that the protons in formaldehyde have the largest known positive geminal coupling constant (41 Hz). I thought only couplings between non-equivalent protons can be seen in the spectrum. Since formaldehyde has C2v symmetry and (if I remember correctly) coupling constants cannot be calculated from theory accurately, how come this coupling is apparent in the spectrum?

• You are right, the normal 1D proton spectrum wouldn't show the coupling. u-of-o-nmr-facility.blogspot.com/2008/06/… I just can't remember whether there is actually an experimental way to see the coupling. – orthocresol Sep 19 '16 at 17:04
• does this help? chem.wisc.edu/areas/reich/nmr/05-hmr-04-2j.htm – MaxW Sep 19 '16 at 19:39
• It's a useful link, but it is still not clear to me whether these are theoretical or experimental. I'd assume this means the values are calculated theoretically but not visible in the spectrum as orthocresol said (as in couplings between equivalent nuclei), but I'm not sure whether that is indeed the case? – Snakeysnakes Sep 19 '16 at 20:17
• Good to see Reich's and Facey's websites referenced as first two comments. They should be prerequisite reading for any [nmr] question here. – long Sep 20 '16 at 1:59

This is an extrapolation of an experimentally measured coupling constant. It is possible to determine the coupling constant, $J_{\ce{H-H}}$ based on the $J_{\ce{H-D}}$, where the following relationship is valid:
$$\left |J_{\ce{H-H}} \right | = (\gamma_\ce{H}/\gamma_\ce{D}) \cdot \left |J_{\ce{H-D}}\right |$$
Shapiro, Kopchick and Ebersole reported the coupling constant for $J_{\ce{H-D}}$ in formaldehyde-d1 in 1963 (The Journal of Chemical Physics 39, 3154 (1963); doi: 10.1063/1.1734163), and calculated the $J_{\ce{H-H}}$ from this. Changes in coupling constants due to the influence of isotopically-modified electron density are apparently very modest for H/D, and so this relationship appears to hold well.
You can actually check this isotope relationship very easily with common NMR solvents and the $\ce{C-H}$ and $\ce{C-D}$ coupling if you run carbon experiments ($\gamma_\ce{H}/\gamma_\ce{D}=6.51439$).
$$\begin{array}{ccc} \hline \text{Solvent} & J_{\ce{C-D}} & J_{\ce{C-H}}\\[1.1em] \hline \ce{CHCl3} & 32 & 208\\ \ce{DMSO} & 21 & 137 \\ \hline \end{array}$$