Skip to main content
. Author manuscript; available in PMC: 2015 Jul 27.
Published in final edited form as: J Am Chem Soc. 2012 Oct 16;134(42):17661–17670. doi: 10.1021/ja3071682

Table 2. Comparison of computed g-values from DFT models with experiment.

gx gy gz
1 Experiment 20052 2.0042 2.0022
DFT model Deviation from experiment [in ppt]
2 Model 4 (with WAT1) -0.3 -0.1 -0.4
3 Model 6 (wirhout WATl) -0.1 -0.1 -0.3
4 graphic file with name nihms415665t1.jpg -0,6 -0.3 -0,5
5 graphic file with name nihms415665t2.jpg -0.2 -02 -0.4
6 graphic file with name nihms415665t3.jpg 0.3 0.2 -0.2
7 graphic file with name nihms415665t4.jpg 0.8 0,1 -0,4
S graphic file with name nihms415665t5.jpg 0.9 0.3 0.0
9 graphic file with name nihms415665t6.jpg 1.7 0,4 0.0

The optimized structure without water (entry 3, model 6) is shown in Figure S5. The optimized structure with water (entry 2, model 4) is shown in Figure 4. Small model systems were obtained by omitting residues of model 4, which are not involved in the PCET pathway. Hydrogen bond lengths are given in Å. Geometries and EPR parameters are calculated using COSMO (ε = 4.0). The uncertainty in the experimental values (entry 9) at 180 GHz is ± 0.1 ppt. The unit ppt refers as part per thousand of the g-value.