Assignment of chemical shifts in benzohydroxamic acid and structure of its silylated derivatives by 15N enrichment

Article abstract of DOI:10.1002/mrc.1217

¹⁵N isotopic enrichment was necessary for the unequivocal assignment of the ¹H NMR lines to the protons in the NH-OH fragment of benzohydroxamic acid, BHXA, C₆H₅CONHOH, in dry dimethyl sulfoxide solutions. The a

Full text of DOI:10.1002/mrc.1217

MAGNETIC RESONANCE IN CHEMISTRY
Magn. Reson. Chem. 2003; 41: 626–628
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/mrc.1217
Spectral Assignments and Reference Data
acid is also a stronger acid than acetohydroxamic acid.10 Hence
Assignment of chemical shifts in
the 1H NMR spectra of benzohydroxamic acid appear to be more
strongly influenced by possible proton exchanges. For these reasons
we decided to test the validity of the assignment.
benzohydroxamic acid and structure of its
silylated derivatives by 15N enrichment
We have already shown1 that the argument according to which
the line with a larger linewidth was assigned to the NH proton4 was
Jan Schraml,1∗ Vratislav Blechta,1 Ludmila Soukupova,
1
´
false as the line was not broadened by quadrupolar effects of the 14
N
1
Jindrich Karban and Jaromı´r Mindl2
ˇ
nucleus of the NH group but by exchange with water present in the
solution.
1
Institute of Chemical Process Fundamentals, Academy of Sciences of
In order to reduce the rates of possible exchange processes, the
concentration of benzohydroxamic acid in dimethyl sulfoxide must
be kept very low (in NMR terms) and the solution must be freshly
prepared from the recrystallized acid. Low concentration and long
T1 relaxation time make assignment experiments8 difficult so we
had to resort to 15N enrichment. Silyl derivatives of the enriched
acid were also used to support the earlier tentative assignment of the
lines in the spectra of disilylated11 and monosilylated12 acids and in
the latter derivatives also to prove their structures.
the Czech Republic, Rozvojova´ 135, 16502 Prague 6, Czech Republic
2
ˇ
´
Department of Organic Chemistry, University of Pardubice, Cs. legiı
565, 53210 Pardubice, Czech Republic
Received 4 February 2003; revised 7 April 2003; accepted 11 April 2003
15N isotopic enrichment was necessary for the unequiv-
ocal assignment of the 1H NMR lines to the protons in
the NH–OH fragment of benzohydroxamic acid, BHXA,
C6H5CONHOH, in dry dimethyl sulfoxide solutions. The
assignment [d.NH/ = 11.21, d.OH/ = 9.01, 1J.15N,1H/ =
102.2 Hz, 2J.15N,1H) <1.5 Hz], which is opposite to that
used by other authors, confirms the assignment extended
to BHXA by Brown and co-workers from the spectra
of acetohydroxamic acid. The enrichment allowed also
assignment of the 29Si lines in the spectra of disilylated
benzohydroxamic acid, (Z)-tert-butyldimethylsilyl N-
tert-butyldimethylsilyloxybenzoimidate (2) and (Z)-tert-
butyldiphenylsilyl N-tert-butyldiphenylsilyloxybenzoi-
midate (3), and confirmed structure of the monosily-
lated products, N-tert-butyldiphenylsilyloxybenzamide
(4) and N-tert-butyldiphenylsilyloxy benzoimidic acid
(5). Copyright  2003 John Wiley & Sons, Ltd.
EXPERIMENTAL
Synthesis
[
15N]Benzohydroxamic acid (1)
A solution of benzoyl chloride (1.92 g, 13.7 mmol) in diethyl ether
°
(3 ml) was added dropwise within 10 min at 0 C under nitrogen to
a stirred mixture of 15NH2OHÐHCl (0.98 g, 14 mmol), K2CO3 (1.93 g,
14 mmol), diethyl ether (10 ml) and water (0.2 ml). After 8 h at room
temperature, the mixture was filtered, concentrated under reduced
pressure and the crystalline residue (0.72 g) was recrystallized from
°
ethyl acetate–heptane. Yield 0.58 g (31%), m.p. 128–129 C. The
°
product was recrystallized from hot water (95 C) twice before NMR
measurements. The reported NMR results were obtained from a
0.03 mM solution of 1 in dry DMSO-d6.
KEYWORDS: NMR; 1H NMR; 13C NMR; 15N NMR; 29Si NMR;
chemical shifts; coupling constants; benzohydroxamic acid;
(Z)-tert-butyldimethylsilyl N-tert-butyldimethylsilyloxy-
benzoimidate; N-tert-butyldiphenylsilyloxybenzamide
(Z)-tert-Butyldimethylsilyl
[
15N]-N-tert-butyldimethylsilyloxybenzoimidate (2)
N-(tert-Butyldimethylsilyl)-N-methyltrifluoroacetamide (209.5 mg,
0.87 mmol) was added to a solution of [15N]-benzohydroxamic acid
(0.03 g, 0.216 mmol) in acetonitrile (0.350 ml) and the mixture was
°
stirred at 80 C for 5 h. The excess of silylating agent and acetoni-
°
trile was removed under reduced pressure (660 Pa, 50 C, 5 h). Yield
INTRODUCTION
43 mg (55%), yellow syrup.
Structural chemistry and spectral studies of hydroxamic acids
and their derivatives have been plagued with errors.1 The errors
include wrong formulae of the synthesized compounds (in the
absence of elemental analysis),2,3 misinterpreted IR spectra of
solid acidic salts1 and contradictory assignments in the 1H NMR
spectra.4–7 The controversial assignment of NH–OH proton lines
in acetohydroxamic acid (in dimethyl sulfoxide solutions) was
unambiguously solved by Brown et al.6 by 15N isotopic enrichment.
With the usual assumption that 1Jꢀ15N,1Hꢁ > 2Jꢀ15N,O–1H), the
high-frequency line (υ D 10.36) was assigned to the NH proton as
it split into a doublet (J D 102.2 Hz) while the other line (υ D 8.69)
remained a singlet upon enrichment. Later, we showed that these
lines in aliphatic hydroxamic acids can be assigned without isotopic
enrichment, e.g. by observing the 13C signal of carbonyl carbon while
selectively decoupling the NH or OH proton.8
(Z)-tert-Butyldiphenylsilyl
[
15N]-N-tert-butyldiphenylsilyloxybenzoimidate (3)
tert-Butyldiphenylsilyl chloride (221 mg, 0.804 mmol) was added
to a solution of [15N]benzohydroxamic acid (0.046 g, 0.335 mmol)
and imidazole (52 mg, 0.737 mmol) in dry N,N-dimethylformamide
°
(2 ml). The mixture was stirred at 78 C for 2 h, left overnight at room
temperature, diluted with water (5 ml) and extracted with diethyl
ether (3 ð 2 ml). The organic phase was dried over Na2SO4 and
concentrated under reduced pressure to yield 190 mg (92%) of crude
crystalline product.
[
[
15N]-N-tert-Butyldiphenylsilyloxybenzamide (4) and
15N]-N-tert-butyldiphenylsilyloxybenzoimidic acid (5)
There were prepared as a mixture12 from [15N]benzohydroxamic
acid according to the procedure described by Muri et al.13 Recrystal-
lization from diethyl ether–hexane yielded 84 mg (64%).
Brown et al.7 extended the validity of their assignment to
.
.
benzohydroxamic acid (BHXA) [i.e. υ(NH) D 11.1 and υ(OH) D 9.0].
However, benzohydroxamic acid differs from aliphatic hydroxamic
acids in a number of properties. The spectra of aliphatic hydroxamic
acids and their N-methyl derivatives indicate the presence of
two isomers6,9 whereas benzohydroxamic acids under comparable
conditions yield only one set of lines in the spectra. Benzohydroxamic
Compounds 2–5 were measured in dry CDCl3 (with 1% HMDSS)
solutions at concentrations varying between 0.1 and 0.4 mM.
Spectra
Solution 1H, 13C, 29Si and 15N NMR spectral measurements were
performed on Varian UNITY-500 and Inova-500 spectrometers
(operating at 499.9 MHz for 1H, at 125.7 MHz for 13C, at 99.309 MHz
for 29Si and at 50.667 MHz for 15N NMR measurements) using
5 mm switchable broadband probes. In all cases the standard
ŁCorrespondence to: Jan Schraml, Institute of Chemical Process
Fundamentals, Academy of Sciences of the Czech Republic, Rozvojova´ 135,
16502 Prague 6, Czech Republic. E-mail: schraml@icpf.cas.cz
Contract/grant sponsor: Grant Agency of the Academy of Sciences;
Contract/grant number: A4072005.
°
software was used. All spectra were recorded at 25 C. The
1H NMR spectra were measured using
a spectral width of
Contract/grant sponsor: Ministry of Education, Youth and Sports of the
Czech Republic; Contract/grant number: CIMSM 253 100001.
8000 Hz and an acquisition time 4 s; FID data were zero-filled
Copyright  2003 John Wiley & Sons, Ltd.

627
Spectral Assignments and Reference Data
to 128 K. The spectra measured in DMSO were referenced to
H
N
H
N
the line of the solvent (υ D 2.500) and spectra measured in
CDCl3 were referenced to internal 1% hexamethyldisilane (HMDSS,
υ D 0.040). The 13C NMR spectra were measured using a spectral
width of 20 000–30 000 Hz. GARP decoupling was applied both
during acquisition (1–2 s) and relaxation delay (2–5 s). Up to
3000 transients were accumulated. Zero-filling to 128 K and mild
line broadening were used in data processing. The spectra were
referenced to the line of the solvent (υ D 39.70 and 76.99 in
DMSO and CDCl3, respectively). The 13C,13C coupling constants
were determined by INADEQUATE experiments14 performed on
unenriched samples of higher concentration. The 29Si,13C coupling
constants were measured15 by 29Si,13C HMQC using INEPT for 29Si
O
SiPh2t-Bu
O
H
O
O
1
6
2
3
5
4
1
4
SiR2t-Bu
OH
O 4
3
1
line enhancement in a dedicated 29Sif H,13Cg pulsed field gradient
15
°
5 mm probe (Nalorac). N NMR spectra were measured using 90
excitation pulses, 60 s relaxation delay and 2 s acquisition for a
spectral width of 30 kHz. Usually, 100 transients yielded spectra
with a sufficient signal-to-noise ratio. The spectra were referenced
externally to the 15N NMR line of nitromethane in 50% solution in the
same deuterated solvent; no susceptibility correction was applied.
The 29Si NMR spectra, referenced to HMDSS at υ D ꢀ19.79, were
measured by the INEPT pulse sequence optimized for trimethylsilyl
groups.16
1
O
N
O
N
2
SiR2t-Bu
SiPh2t-Bu
2 R = CH3
3 R = C6H5
5
Scheme 1
The silicon atoms in 2 and 3 are labeled according to the label of the
oxygen atom to which they are bonded (Si-1 and Si-4).
RESULTS AND DISCUSSION
The splitting of the 1H NMR line at υ 11.21 upon 15N enrichment
into a doublet (with a coupling constant of 102.2 Hz) confirms
validity of Brown et al.’s assignment for benzohydroxamic acid. The
two NH proton lines exhibit a further splitting of 1.8 Hz which was
proven by selective decoupling of the proton line at υ 9.01 to be due to
The experimental results for 1–5 are summarized in Table 1 and their
structures are shown in Scheme 1. The indicated Z geometry around
the C N bond has been proven11 only for 2 by observing 1Jꢀ13C,13C)
coupling of the C N carbon (75.0 Hz). The stereochemistry for 3 is
proven similarly in the present study but for 5 it is tentative only.
Table 1. Chemical shifts and coupling constants in derivatives of benzohydroxamic/benzohydroximic acida
υꢀ1H) [Jꢀ15N,1H)]
υꢀ13C) [Jꢀ15N,13C)]
C-3/5 C-4
υꢀ29Si)
Compound
NH
OH
9.01b
[Jꢀ15N,29Si)] υꢀ15N)
C
O/C
N
C-1
C-2/6
CH3Si CH3 –(C)
—
C–(Si)
—
1
11.21b
—
ꢀ216.0
ꢀ76.8
ꢀ77.6
ꢀ214.2
ꢀ83.8
164.39
(13.7)
133.00 127.07 128.58 131.34
(11.2) (0.5)
133.29 126.35 128.00 129.38
(10.6) (3.1)
132.86h 126.64
(102.2) (<1.5)
2
3
4
5
—
—
—
—
27.29 (1.8)c
23.13 (0.6)d
152.96e
ꢀ3.18;ꢀ4.93f
19.00;18.61
20.26;19.27j
19.21
(<0.6)
26.47;25.97
i
i
i
l
—
2.67 (1.4)c
ꢀ3.34 (0.6)d
151.65g
—
—
—
—
27.66;26.48
26.77
(<0.3)
(10.6)
(3.2)
i
i
7.92
6.44
166.8k
128.91l
(9.2)
—
—
—
(93.9)
(0.4)
i
i
—
8.31
2.22
(1.3)
158.82m
132.26l
(11.9)
—
27.27
19.60
(<1.5)
(<0.6)
a Chemical shift values (ppm) are on the υ scale; for the details (solvents, concentrations and temperature), see Experimental. 15N coupling
constants in parentheses are the absolute values in Hz.
b 3Jꢀ1H,N–O–1Hꢁ D 1.8 Hz.
c Si-1.
d Si-4.
e 1Jꢀ13C,13Cꢁ D 75.0 Hz.
f 3Jꢀ15N,O–Si–13Cꢁ D 1.5 Hz.
g 1Jꢀ13C,13Cꢁ D 75.7 Hz; Jꢀ13C,29Si-1ꢁ D 4.9 Hz, Jꢀ13C,29Si-4ꢁ D 4.6 Hz.
h Jꢀ13C,29Si-1ꢁ < 0.5, Jꢀ13C,29Si-4ꢁ D 2.3 Hz.
i Not assigned.
j 3Jꢀ15N,O–Si–13Cꢁ D 0.4 Hz estimated from linewidth.
k Broad line, Jꢀ13C,29Siꢁ D 1 Hz.
l Jꢀ13C,29Siꢁ < 0.5 Hz.
m Jꢀ13C,29Siꢁ D 3.9 Hz.
Copyright  2003 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2003; 41: 626–628

628
Spectral Assignments and Reference Data
coupling with the OH proton. No such splitting is visible on the OH
proton line because of exchange broadening of the line (linewidth
4.1 Hz). The exchange also obscures any coupling of the OH proton
with the 15N nucleus. The 1Jꢀ15N,1H) coupling in 4 is considerably
smaller than in 1.
9.01 is due to the OH proton. It is the latter which exchanges faster
with protons of water if present in the solution. The two disilylated
benzohydoxamic acids 2 and 3 have hydroximic structures and
both are Z isomers. Their 29Si lines assigned according to the
chemical shift values exhibit coupling with 15N nuclei, the coupling
constants following the relation 2Jꢀ15N,O–29Siꢁ > 3Jꢀ15N,C–O–29Si).
The 15N chemical shifts found here agree very well with those
determined without enrichment in considerably more concentrated
solutions;11,12 the shift for 3 has not been reported before. The shifts
in the range ꢀ210 to ꢀ220 confirm the hydroxamic nature of 1 and 4
and the shifts in the range ꢀ70 to ꢀ90 demonstrate the hydroximic
structure of 2, 3 and 5.1
Monosilylation of BHXA produces a mixture of 4 and 5. The 29Si,15
N
coupling supports the suggested structure of 5, but in the absence
of 1Jꢀ13C,13C) coupling the Z configuration must be considered to be
only tentative.
The 29Si NMR spectra, in addition to reproducing the already
reported chemical shifts (for 2, 4 and 5),11,12,17 provided the
values of 29Si,15N coupling constants. In the disilylated derivative
2 the lines at υ 27.29 and 23.13 have been tentatively assigned
earlier to Si-1 and Si-4.17 This assignment leads to a relation
2Jꢀ15N,O–29Siꢁ > 3Jꢀ15N,C–O–29Si), but there are in sufficient data in
the literature to confirm the validity of this relationship. Although the
29Si chemical shifts in 3 differ substantially from those in 2 (owing to
substitution of two methyl groups by phenyl groups), all the coupling
constants of the silicon nuclei confirm the assignment of the signal
at υ 2.67 to Si-1 and that at υ ꢀ3.34 to Si-4. The 1Jꢀ13C,13C) coupling
constant between the C N and aromatic Cipso carbon atoms confirms
the Z arrangement around the C N bond in 3 on the basis of the
same argument as used for the other disilylated benzohydroximic
acids studied earlier.11 In 5, both the silicon chemical shift and its
coupling with 15N have values very close to those found for Si-1 in
3, thus confirming the proposed structure of 512 and suggesting a
similar Z arrangement around the C N bond as found in 2 and 3.
Further support for this conclusion was sought in the measurements
of 13C,13C coupling constants, but these experiments failed because
of the low solubility of the mixture of 4 and 5 at room temperature
and decomposition at a higher temperature.
Acknowledgments
The authors gratefully acknowledge financial support provided by
the Grant Agency of the Academy of Sciences (grant No. A4072005)
and Ministry of Education, Youth and Sports of the Czech Republic
(Research Project CIMSM 253 100001). Numerous illuminating
discussions with Professor Otto Exner are greatly appreciated.
REFERENCES
1. Schraml J. Appl. Organomet. Chem. 2000; 14: 604.
2. Hauser CR, Renfrow WB Jr. Org. Synth. Coll. Vol. 1959; 2: 67.
´
3. Podlaha J, Cısarˇova´ I, Soukupova´ L, Schraml J, Exner O. J. Chem.
Res. (S) 1999; 520.
4. Exner O. Dan. Tidsskri. Farm. 1968; 42: 145.
5. Shendrovich VA,
Vajnshenker AI, Dogadina AV. Zh. Obshch. Khim. 1979; 49: 1746.
6. Brown DA, Glass WK, Mageswaran R, Girmay B. Magn. Reson.
Chem. 1988; 26: 970.
7. Brown DA, Coogan RA, Fitzpatrick NJ, Glass WK, Abuk-
shima DE, Shiels L, Ahlgre´n M, Smolander K, Pakkanen TT,
Pakkanen TA, Pera¨kyla¨ M. J. Chem. Soc., Perkin Trans. 2 1996;
2673.
Ryaboj VI,
Kriveleva ED,
Ionin BI,
The lines of aromatic carbons in the NMR spectra 1 and
2 were assigned by INADEQUATE18 experiments performed on
concentrated solutions and some lines in 3, 4 and 5 were assigned
by analogy. It is surprising to see that whereas C-1 coupling with
15N [2Jꢀ15N,C–13C)] varies only slightly in the studied compounds,
the coupling over one bond is observable only in 1, but it is not
resolved in the spectra of other compounds.
´
8. Schraml J, Kvıcˇalova´ M, Blechta V, Soukupova´ L, Exner O,
Boldhaus H-M, Erdt F, Bliefert C. Magn. Reson. Chem. 2000; 38:
795.
9. Brown DA, Glass WK, Mageswaran R, Mohammed SA. Magn.
Reson. Chem. 1991; 29: 40.
Results obtained for 4 are difficult to interpret as there are
no available data for comparison. The compound clearly has the
hydroxamic structure (see 15N chemical shift), the 29Si chemical
shift is to a higher positive υꢀ29Si) value from that in 5 in line with
the silicon atom being attached to an N–O oxygen atom but its
2Jꢀ29Si,O–15N) coupling is much smaller than the same coupling in
the compounds with a hydroximic structure (2 and 3), it is even
smaller than the 3Jꢀ29Si,O–C–15N) couplings in these compounds.
10. Menon SK, Agrawal YK. Rev. Inorg. Chem. 1996; 16: 1.
´
11. Schraml J, Kvıcˇalova´ M, Soukupova´ L, Blechta V. Magn. Reson.
Chem. 1999; 37: 427.
12. Schraml J, Soukupova´ L, Blechta V, Karban J, Cısarˇova´ I. J.
´
Organomet. Chem. 2001; 628: 81.
13. Muri D, Bode JW, Carreira EM. Org. Lett. 2000; 2: 539.
14. Bax A, Freeman R, Kempsell SP. J. Am. Chem. Soc. 1980; 102: 4849.
15. Blechta V, Sy´kora J, Schraml J. In 18th NMR Valtice Proceedings,
Valtice, Czech Republic, 2003; 17.
Also, 15N couplings of 13C nuclei appear anomalous. While the C
O
carbon clearly has the shift of a carbonyl (and not of C N carbon),
no coupling with the 15N nucleus was observed, and coupling of the
C-1 phenyl carbon is the lowest among the studied compounds. We
cannot exclude the possibility that some exchange process, involving
a nitrogen atom, is taking place in chloroform solutions of 4.
16. Schraml J. Collect. Czech. Chem. Commun. 1983; 48: 3402.
´
17. Schraml J, Kvıcˇalova´ M, Soukupova´ L, Blechta V, Exner O. J.
Organomet. Chem. 2000; 597: 200.
´
18. Schraml J, Kvıcˇalova´ M, Soukupova´ L, Blechta V, Exner O. J.
Phys. Org. Chem. 1999; 12: 668.
CONCLUSIONS
The high-frequency line (υ 11.21) in the 1H NMR spectrum of BHXA
(1) is unequivocally assigned to the NH proton and the line at υ
Copyright  2003 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2003; 41: 626–628