Comparison of the 13C (C = N) Chemical Shifts of XArC(Me) = NArY and XArCH = NArY
added. Then the mixture was heated at 140° Celsius for 10 h, cooled to
room temperature, and purified by recrystallization or column chroma-
tography. The obtained products were characterized by NMR.
compounds with the same pairs of substituents X and Y were
employed. It was surprising that there is no linear relationship
of the δC(C = N) values of XBAYs versus that of XPEAYs.
Based on Fig. 1 and Eqn (2), we try to add the excited-state
substituent parameters σ of X and Y[26] into Eqn (2); then Eqn
ex
Spectroscopic measurement
cc
ex
cc
ex
cc
(3) was obtained by adding σ (X) and σ (Y).
The products were vacuum dried for a whole day before measurements.
The NMR spectra of compounds XPEAYs in Table 1 were recorded in
CDCl3 at 293 K. The 13C NMR chemical shifts are expressed in ppm rela-
tive to CDCl3 (77.00 ppm) at 126 MHz, and the 1H NMR chemical shifts
are expressed in ppm relative to TMS (0.00 ppm) at 500 MHz. All samples to
be measured were made up at an appropriate concentration (32 mg/ml CDCl3).
The 1H NMR and 13C NMR spectral date and spectrum are given in the
Supporting Information.
δCðC ¼ NÞ ¼ 165:43 – 2:80σFðXÞ þ 2:06σFðYÞ þ 0:04σRðXÞ
þ0:19σRðYÞ–0:25Δσ2–0:09σeccxðXÞ þ 0:38σcecxðYÞ
R ¼ 0:9907; R2 ¼ 0:9814; S ¼ 0:13; n ¼ 61; F ¼ 399:44 (3)
It was observed that Eqn (3) has better correlation than Eqn (2);
the standard error was 0.13 ppm, which is within the experimental
uncertainties. It indicates that the excited-state substituent para-
meter of substituents X and Y play a role of affecting the δC(C = N)
of XPEAYs. In addition, there is a group Me attached to the C = N
bond in XPEAYs. Because the C = N bond is a polar double bond,
the electronegativity of N atom is bigger than that of C atom; the
π-electron flow from C to N. Maybe the electron-donating group
Me affects the δC(C = N) by means of the substituent specific cross-
interaction effect between Me and Y. So we further put this item,
namely χ2 =[σP(Me)ꢀσ(Y)]2 =[ꢀ0.17ꢀσ(Y)]2 into Eqn 3 and then
obtained Eqn (4).
It is known that the concentration of a compound will influence its
13C NMR chemical shifts. In order to investigate this effect, we choose
p-O2NPEAMe-p as an example to measure its δC(C = N) at a gradient
concentration of 30, 35, and 50 mg/ml in the solvent of CDCl3. The
δC(C = N) are 163.57, 163.59, and 163.57 ppm, respectively, which are all
close to the value 163.54 ppm measured at concentration of 32 mg/ml in
CDCl3. It is to say that a small change of the concentration has unobvious
effect on the δC(C = N) values for the target compounds XPEAYs.
In addition, different values of δC(C = N) for a compound will be ob-
tained in a variety of solvent. Thus, we choose some XPEAYs as model
compounds (e.g., HPEACN-m, HPEAMeO-p, p-F3CPEAMeO-p, p-O2NPEAH)
to measure their δC(C = N) in the solvent DMSO-d6. The obtained δC(C = N)
values are 166.88, 164.96, 164.16, and 164.12 ppm, respectively, which
matches the variation trend of δC(C = N) measured in CDCl3.
δCðC ¼ NÞ ¼ 165:43– 2:76σFðXÞ þ 1:74σFðYÞ þ 0:08σRðXÞ
ꢀ0:01σRðYÞ ꢀ 0:34Δσ2 ꢀ 0:1σecxcðXÞ þ 0:48σcexcðYÞ
þ0:42χ2
METHODS AND DISCUSSION
Cao[18] proposed Eqn (1) to correlate the δC(C = N) of XBAYs, in
R ¼ 0:9931; R2 ¼ 0:9862; S ¼ 0:11; n ¼ 61; F ¼ 466:69 (4)
which the standard error is only 0.17 ppm.
δCðC ¼ NÞ ¼ 160:25 ꢀ 4:18σFðXÞ þ 3:24σFðYÞ ꢀ 1:15σRðXÞ
þ4:67σRðYÞ –0:59Δσ2
The result of Eqn (4) has better improvement than Eqn (3), and
its standard error is down to 0.11 ppm. What we also want to
know if there is substituent specific cross-interaction effect
between Me and substituent X. Thus, we added a corresponding
item, ω2 = [σP(Me)ꢀσ(X)]2, into Eqn (3) and carried out a regres-
sion analysis. The obtained result shows that the effect of ω2
on the δC(C = N) is very little and can be ignored.
R ¼ 0:9975; R2 ¼ 0:9950; S ¼ 0:17; n ¼ 80; F ¼ 2937:31 (1)
Where σF(X) and σF(Y) are the inductive parameters of substit-
uents X and Y, σR(X), and σR(Y) are the conjugative parameters of
X and Y, respectively, Δσ2 is the substituent specific cross-
interaction effect between substituents X and Y.
In order to compare with the compounds XBAYs, here, 61
samples of XPEAYs were synthesized; their 13C NMR of C = N
bonds chemical shifts were measured, which were listed in
Table 1. Firstly, the variants of Eqn (1) were employed to quantify
the δC(C = N) of XPEAYs. Thus, the experimental values of the
δC(C = N) of XPEAYs in Table 1 were regressed against the five
parameters σF(X), σF(Y), σR(X), σR(Y), and Δσ2 according to Eqn (1),
and Eqn (2) was obtained.
By analyzing the t-values (Table 2) of the coefficients of Eqn (4),
we can see that the contributions of items σR(X), σR(Y), and σeccx(X)
are relatively small. Thus, we remove these three items, that is,
2
2
ex
cc
using the five items σF(X), σF(Y), Δσ , σ (Y), and χ to carry out
regression analysis again. Equation (5) was obtained.
δCðC ¼ NÞ ¼ 165:41– 2:67σFðXÞ þ 1:77σFðYÞ ꢀ 0:29Δσ2
ex
þ0:50σ ðYÞ þ 0:39χ2
cc
R ¼ 0:9922; R2 ¼ 0:9844; S ¼ 0:12; n ¼ 61; F ¼ 692:06 (5)
δCðC ¼ NÞ ¼ 165:47 – 2:71σFðXÞ þ 2:19σFðYÞ þ 0:26σRðXÞ
þ0:60σRðYÞ ꢀ 0:42Δσ2
It can be seen that the results of Eqn (5) rival to that of Eqn (4),
in spite of the three parameters, were reduced in the former.
Equation (5) shows that there are five main factors affecting
the δC(C = N) of XPEAYs, which include both inductive effects of
substituents X and Y (σF(X) and σF(Y)), the excited-state substitu-
R ¼ 0:9784; R2 ¼ 0:9573; S ¼ 0:20; n ¼ 61; F ¼ 246:90 (2)
ex
ent parameter of Y (σ (Y)), the two substituent specific cross-
Equation (2) has also good correlation, because the molecular
structure of XPEAYs has certain similarity with that of XBAYs. Its
standard error 0.20 ppm is larger than that of Eqn (1)[18]; maybe
there are some others factors affecting the δC(C = N) of XPEAYs.
To seek out this difference, we plot the experimental δC(C = N)
of XBAYs against that of corresponding XPEAYs (e.g., p-MeBAF-p
versus p-MePEAF-p), and obtained Fig. 1, in which 45 couples of
cc
interaction effects between X and Y (Δσ2) and between Me and Y
(χ2). The factors affecting the δC(C= N) of XPEAYs are quite diffe-
rent from that of XBAYs. Hence, we recommended Eqn (5) to quan-
tify the δC(C= N) of XPEAYs. The values calculated by Eqn (5) were
listed in the last column of Table 1. Figure 2 is the plot of calculated
values of δC(C= N) versus the experimental ones.
J. Phys. Org. Chem. 2015, 28 564–569
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc