Assessing the proton affinities of N, N ′-diamidocarbenes

Article abstract of DOI:10.1021/jo401902c

The gas-phase proton affinities (PAs) of a series of novel diamidocarbenes (DACs) were assessed and compared to various imidazolylidene-based N-heterocyclic carbenes (NHCs) through experimental and computational methods. Apart from a perfluorinated-phenyl derivative (PA = 233 kcal/mol), the calculated and measured PAs for a range of DACs (256-261 kcal/mol) were comparable to those of the NHCs (260-266 kcal/mol). Proton transfer from the protonated carbene to various reference bases, as observed by mass spectrometry, was inhibited by steric bulk and precluded the direct measurement of the PA for the known DACs, N,N′-dimesityl-4,6-diketo-5,5-dimethylpyrimidin-2-ylidene and N,N′-diisopropylphenyl-4,6-diketo-5,5-dimethylpyrimidin-2-ylidene. However, DACs featuring less hindered N-aryl substituents facilitated proton transfer, and the measured PA values were found to be consistent with density functional theory calculations (B3LYP/6-31+G(d)). Notably, the PAs of the DACs studied were similar to those of the NHCs, indicating that the former retain many of the nucleophilic characteristics intrinsic to their parent diaminocarbenes and that the observed differences in chemical reactivity may be primarily attributed to an enhanced electrophilicity.

Full text of DOI:10.1021/jo401902c

Article
pubs.acs.org/joc
Assessing the Proton Affinities of N,N′‑Diamidocarbenes
Mu Chen,† Jonathan P. Moerdyk,‡ Garrett A. Blake,‡ Christopher W. Bielawski,*,‡ and Jeehiun K. Lee*,†
†Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, New Brunswick, New Jersey 08901,
United States
‡Department of Chemistry and Biochemistry, The University of Texas at Austin, Austin, Texas 78712, United States
S
* Supporting Information
ABSTRACT: The gas-phase proton affinities (PAs) of a series of
novel diamidocarbenes (DACs) were assessed and compared to
various imidazolylidene-based N-heterocyclic carbenes (NHCs)
through experimental and computational methods. Apart from a
perfluorinated-phenyl derivative (PA = 233 kcal/mol), the calculated
and measured PAs for a range of DACs (256−261 kcal/mol) were
comparable to those of the NHCs (260−266 kcal/mol). Proton
transfer from the protonated carbene to various reference bases, as
observed by mass spectrometry, was inhibited by steric bulk and
precluded the direct measurement of the PA for the known DACs,
N,N′-dimesityl-4,6-diketo-5,5-dimethylpyrimidin-2-ylidene and N,N′-diisopropylphenyl-4,6-diketo-5,5-dimethylpyrimidin-2-yli-
dene. However, DACs featuring less hindered N-aryl substituents facilitated proton transfer, and the measured PA values were
found to be consistent with density functional theory calculations (B3LYP/6-31+G(d)). Notably, the PAs of the DACs studied
were similar to those of the NHCs, indicating that the former retain many of the nucleophilic characteristics intrinsic to their
parent diaminocarbenes and that the observed differences in chemical reactivity may be primarily attributed to an enhanced
electrophilicity.
results in catalysts with superior activities,12−14 and react with
INTRODUCTION
■
elemental sulfur and electrophilic heteroallenes.15
Traditional carbenes (e.g., methylene) are typically electrophilic
Recently, we developed the N,N′-diamidocarbenes (DACs,
Figure 1) which are not only stable but feature enhanced
electrophilic properties due to the strategically placed carbonyl
and display commensurate chemical reactivity profiles including
abilities to insert into C−H bonds, activate H2, cyclopropanate
olefins, and couple to carbon monoxide to form ketenes.1−4
groups that draw electron density from the adjoining nitrogen
atoms away from the carbene center.16−20 The reactivity profile
However, the transiency of methylene and other electrophilic
carbenes also requires in situ generation methods, which can be
of the DACs is similar to that of traditional carbenes, and
examples of C−H insertions,16 CO fixation,16,17,21 and NH3
problematic and avoided through the use of isolable analogues.
activation17 have been reported. Using computational methods,
Following the first report of a stable carbene by Bertrand over
two decades ago,5 Arduengo and co-workers isolated the first
the unique reactivities of the DACs have been attributed to the
crystalline carbene through incorporation into an N-hetero-
cyclic scaffold.6 The carbenoid character (or lack thereof) of the
N-heterocyclic carbenes (NHCs, Figure 1) has been a topic of
some controversy.7−10 Regardless, when compared to tradi-
tional carbenes, NHCs are relatively nucleophilic and often
display different reactivities. For example, NHCs do not
activate H2 nor fix carbon monoxide to make ketenes;11 they
do, however, bind to transition metals, a process that often
relatively low-lying lowest unoccupied molecular orbital
(LUMO) in conjunction with a highest occupied molecular
orbital (HOMO) energy level similar to that of prototypical
NHCs.22 Thus, while the DACs are relatively electrophilic, they
are still nucleophilic and coordinate to a broad range of
transition metals as well as couple to a range of electrophilic
compounds.16,23,24 Since basicity often correlates with nucleo-
philicity, we reasoned that measuring the proton affinity (PA)
of the DACs would facilitate quantitative comparisons to other
carbenes and guide the development of new classes of DACs
and potentially other stable, electrophilic carbenes.
Several strategies have been developed to study the affinities
of carbenes for protons, including both pKa and PA
measurements. The former largely involve monitoring the
exchange of isotopes in water25−28 or the reaction of the free
Figure 1. General structures for an N-heterocyclic carbene (NHC)
and an N,N′-diamidocarbene (DAC).
Received: August 27, 2013
Published: October 3, 2013
© 2013 American Chemical Society
10452
dx.doi.org/10.1021/jo401902c | J. Org. Chem. 2013, 78, 10452−10458

The Journal of Organic Chemistry
Article
Figure 2. Calculated proton affinities in kcal/mol at 298 K for selected NHCs and DACs using B3LYP/6-31+G(d).
Scheme 1
carbene with hydrocarbon indicators in THF or DMSO.29−31
While these methods enable solution-state determination,
solvent incompatibility can arise, and the resultant values are
highly solvent dependent. In comparison, proton affinity values
are intrinsic to the chemical entity and may be measured in the
gas phase through mass spectrometry.32,33 Recently, we
reported the reaction of reference bases with cationic carbene
precursors in the gas phase which enabled the bracketing of PA
values for the corresponding carbenes.34,35
precursors necessitated a modified introduction method.
Fortunately, electrospray ionization of the formally hydrated
version of the DAC (e.g., 1·H2O; Scheme 1, R = 2,6-
diisopropylphenyl) dissolved in formic acid solution was found
to successfully yield 1H+ as the major signal. The synthesis of
the water adduct of 1 (and other derivatives reported in this
paper) was carried out as shown in Scheme 2.17
Scheme 2
In this paper, we describe experimental and computational
PA studies of known as well as new DACs in the gas phase. We
found that, despite the incorporation of electron-withdrawing
carbonyl groups adjacent to the nitrogen atoms, the DACs
display PA values similar to the NHCs studied herein.
Collectively, these results, which are consistent with density
functional theory calculations and observed chemical reactivity,
indicate that the DACs display similar basicities and
nucleophilic characteristics as the NHCs and provide a means
to quantitatively compare these classes of stable carbenes.
As summarized in Table 1, 1H+ was exposed to various
strong bases in the gas phase to test for the occurrence of
proton transfer. Given the calculated PA of DAC 1 (258.1 kcal/
mol), we were surprised that bases as strong as 2-tert-
butylimino-2-diethylamino-1,3-dimethylperhydro-1,3,2-diaza-
phosphorine (BEMP; PA = 263.8 kcal/mol) failed to
deprotonate 1H+. Similarly, attempts to bracket the PA of
DAC 2 were also unsuccessful. As we had previously bracketed
the PAs of NHCs 3a and 3b (which have relatively small
substituents) experimentally through analogous gas-phase
measurements,34,35 sterics were envisioned to inhibit proton
transfer. Subsequent efforts were directed toward two bulkier
NHC derivatives, 3c and 3d. For 3cH+, we observed proton
transfer to BEMP (PA = 263.8 kcal/mol) but not to tert-
octyliminotris(dimethylamino)phosphorane (tOctP1(dma); PA
RESULTS AND DISCUSSION
■
Our initial efforts were directed toward determining the PAs of
the known DACs 1 and 2 which were calculated by density
functional theory to be similar (ca. 258 kcal/mol); by
comparison, the calculated PA values of the NHCs 3a−d
were only modestly higher (260−266 kcal/mol) than those of
the DACs (Figure 2).36 To experimentally measure the PAs of
the 1 and 2, attention shifted to mass spectrometry and proton-
exchange reactions between the protonated carbene precursors
(i.e., 1H+ and 2H+) and various bases in the gas phase which
had previously been utilized for determining the PA values of
various NHCs.34,35,37 Whereas the prior studies of the NHCs
had been conducted directly with the air-stable salt precursors,
the tendency to form covalent rather than ionic carbene
precursors as well as the high water sensitivity of the DAC
10453
dx.doi.org/10.1021/jo401902c | J. Org. Chem. 2013, 78, 10452−10458

The Journal of Organic Chemistry
Article
kcal/mol) similar to that of 1 and 2 was envisioned (i.e., 5). In
a manner analogous to 1·H2O, the formally hydrated derivative
(i.e., 5·H2O) was synthesized. Electrospray of a formic acid
solution of 5·H2O and the gas-phase introduction of reference
bases to bracket the PA value were explored. Reference bases
with PA values of 260.6 kcal/mol and higher effectively
deprotonated 5H+; in contrast, proton transfer was not
observed for reference bases with PAs below 260.6 kcal/mol.
Instead, when the base HP1(dma) (PA = 257.4 kcal/mol), 7-
methyl-1,5,7-triazabicyclo[4.4.0]dec-5-ene (MTBD; PA = 254.0
kcal/mol), or 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU; PA =
250.5 kcal/mol) was introduced, ions with mass-to-charge
ratios corresponding to the adducts of the protonated DAC and
the reference base (or, alternatively, the protonated reference
base and the DAC) were observed. This result was consistent
with the susceptibility of DACH+ to react with nucleophiles
such as water and the splitting of N−H bonds by free DACs,17
though the actual nature of the complex (e.g., a covalent versus
a noncovalent ion−molecule adduct) has not been probed.
Regardless of its structure, proton transfer cannot be
ascertained when such complexation occurs. However, a lack
of proton transfer with N,N,N′,N′-tetramethyl-1,3-propanedi-
amine (PA = 247.4 kcal/mol) as the reference base did indicate
that the PA of the tolyl DAC 5 was between 247.4 and 260.6
kcal/mol (Table 3). Thus, unlike with DACs 1 and 2, the PA of
Table 1. Summary of the PA Bracketing Results for DACs 1,
2 and NHCs 3c, 3d
b
proton transfer to reference base
a
39
,
ref base
BEMP
PA (kcal/mol)
1H+
2H+
3cH+
3dH+
263.8 2.0
262.0 2.0
260.6 2.0
257.4 2.0
−
−
−
−
−
−
−
−
+
−
−
−
−
tOctP1(dma)
tBuP1(dma)
HP1(dma)
−
−
−
a
BEMP = 2-tert-butylimino-2-diethylamino-1,3-dimethylperhydro-
1,3,2-diazaphosphorine; tOctP1(dma) = tert-octyliminotris-
(dimethylamino)phosphorane; tBuP1(dma) = tert-butylimino-tris-
(dimethylamino)phosphorane; HP1 (dma)
= iminotris-
b
(dimethylamino)phosphorane. The “+” symbol indicates the
observation of proton transfer, and the “−” symbol indicates the
lack of observed proton transfer.
= 262.0 kcal/mol; Table 1). Thus, the PA of 3c was determined
to be between 262.0 and 263.8 kcal/mol, consistent with the
computationally predicted PA of 262.9 kcal/mol. BEMP did
not effect proton transfer with 3dH+ (Table 1), and reference
bases with PAs higher than BEMP were not practical due to
their nonvolatility.38 We therefore assigned the PA for the
corresponding carbene as >263.8 kcal/mol, in agreement with
the calculated value of 265.6 kcal/mol.
In light of our success with the PA measurement of the
NHCs,34,35 we re-examined potential steric inhibition of proton
transfer for DACs 1 and 2. In addition to the relatively large N-
aryl substituents of the aforementioned DACs, the strong bases
needed for our studies were sterically encumbered, which may
further inhibit proton transfer.40−42 Efforts were thus directed
toward the synthesis and study of the perfluorophenyl DAC
derivative 4 (calculated PA = 233.0 kcal/mol) as this substrate
should be less bulky than DACs 1 and 2, and the corresponding
conjugate acid should be more acidic. As such, less basic and
less sterically demanding reference bases would be needed to
bracket the PA of this carbene.
Table 3. Summary of Results for Proton Affinity Bracketing
of Tolyl DAC 5
PA
(kcal/mol)
Proton transfer (between 5H+
a
ref base39,43
and ref base)
BEMP
263.8 2.0
262.0 2.0
260.6 2.0
247.4 2.0
+
+
+
−
tOctP1(dma)
tBuP1(dma)
N,N,N′,N′-tetramethyl-1,3-
propanediamine
a
The “+” symbol indicates the observation of proton transfer, and the
“−” symbol indicates the lack of observed proton transfer.
Using the aforementioned mass spectrometric methodology,
the PA of 4 was measured. While N-methylpiperidine (PA =
232.1 kcal/mol did not deprotonate 4H+, proton transfer was
the less sterically hindered tolyl DAC 5 was successfully
bracketed, indicating that the conjugate acids of DACs 1 and 2
feature sterically inaccessible protons under the conditions
studied.
observed when triethylamine (PA = 234.7
kcal/mol) was
used as the reference base (Table 2). The experimental PA of 4
was therefore established to be 233 kcal/mol, in agreement
with the calculated value of 233.0 kcal/mol.
To further probe how aryl substitution influences the PAs of
the DACs, attention shifted to the anisidyl series 6, as installing
a methoxy group was calculated to increase the PA of the DAC,
with a relative PA order of 6a > 6c > 6b. The DACs 6 were
synthesized in an analogous manner to the other formally
hydrated DAC adducts (Scheme 2) and then subjected to
electrospray analysis and PA bracketing (Table 4). For the
relatively basic DAC 6a, tOctP1(dma) (PA = 262.0 kcal/mol)
effected deprotonation of 6aH+ although tBuP1(dma) (PA =
260.6 kcal/mol) did not. The experimental PA for 6a was
therefore bracketed between 260.6 and 262.0 kcal/mol,
consistent with the calculated value (261.0 kcal/mol). Proton
transfer from the protonated, yet less basic, meta and para
derivatives 6b and 6c was observed with tBuP1(dma) (PA =
260.6 kcal/mol; Table 4, third row). As with the tolyl DAC 5,
the m/z ratios indicated adduct formation between the
protonated reference base and the DAC when employing
reference bases HP1(dma) and MTBD. The lack of proton
transfer with N,N,N′,N′-tetramethyl-1,4,-butanediamine (PA =
250.1 kcal/mol) placed the PA values of 6b and 6c above 250.1
Building upon the aforementioned results obtained for 4, we
sought to expand the PA measurements to include relatively
basic DAC derivatives. Considering the lack of proton transfer
with 1 and 2, a less sterically encumbered DAC featuring a 4-
methylphenyl N-substituent with a calculated PA value (256.3
Table 2. Summary of Results for Proton Affinity Bracketing
of Perfluorophenyl DAC 4
PA
(kcal/mol)
proton transfer (between
a
ref base43
protonated 4H+ and ref base)
N,N-
235.1 2.0
+
dimethylcyclohexylamine
triethylamine
234.7 2.0
232.1 2.0
230.8 2.0
228.0 2.0
+
N-methylpiperidine
N-methylpyrrolidine
piperidine
−
−
−
a
The “+” symbol indicates the observation of proton transfer, and the
“−” symbol indicates the lack of observed proton transfer.
10454
dx.doi.org/10.1021/jo401902c | J. Org. Chem. 2013, 78, 10452−10458

The Journal of Organic Chemistry
Article
NHCs. Thus, we conclude that the observed reactivity
differences between the two classes of carbenes may be
attributed to the enhanced electrophilic characteristics of the
DACs, as evidenced by their relatively low-lying LUMOs and
abilities to react with a broad range of nucleophiles.16,17,44,45
Moreover, having ascertained the PAs of various DACs, the
results described above may be utilized to determine the
compatibility of DACs with various acidic functional groups
(e.g., electron-deficient terminal alkynes or ketones)22,44,45 as
well as to contribute to the design and understanding of new
carbene scaffolds.
Table 4. Summary of Proton Affinity Bracketing Results for
the Anisidyl-Based DACs 6
proton transfer (between
a
protonated 6H+ and ref base)
PA
ref base39,43
BEMP
(kcal/mol)
6a
6b
6c
263.8 2.0
262.0 2.0
260.6 2.0
257.4 2.0
254.0 2.0
+
+
−
+
+
+
+
+
+
tOctP1(dma)
tBuP1(dma)
HP1(dma)
complex
complex
−
complex
complex
−
complex
complex
−
b
MTBD
N,N,N′,N′-tetramethyl- 250.1 2.0
1,4-butanediamine
EXPERIMENTAL SECTION
■
a
The “+” symbol indicates the observation of proton transfer, and the
For NHCs 3a, 3b, and 3d, the corresponding precursors (1,3-
dimethylimidazolium tetrafluoroborate, 1-ethyl-3-methylimidazolium
tetrafluoroborate, and 1,3-diisopropylimidazolium tetrafluoroborate)
were used as received. The precursor to 3c, 1-methyl-3-isopropylimi-
dazolium iodide, was synthesized according to a literature procedure.46
Bis(4-methylphenyl)formamidine,47 bis(2-methoxyphenyl)-
formamidine,48 bis(perfluorophenyl)formamidine,49 2-chloro-1,3-bis-
(2,6-diisopropylphenyl)-4,6-diketo-5,5-dimethylpyrimidine (1·HCl),17
and the formally hydrated carbene adduct 2·H2O17 were synthesized
according to literature procedures. All other commercial substrates
were used as received. All reference bases were used as received.
To generate the protonated carbene species via electrospray, the
formally hydrated DACs were dissolved in formic acid or a mixture of
formic acid/water (1:10 v/v) ([DAC]0 ∼10−4 M). Standard Schlenk
techniques under an atmosphere of nitrogen or in a nitrogen-filled
glovebox were used for the condensation reactions needed to prepare
1,2,4−6·H2O. Benzene and dichloromethane were dried and degassed
using a solvent purification system. Chemical shifts (δ) are reported in
ppm relative to the residual chloroform (1H: 7.24 ppm, 13C: 77.0
ppm) as reference. Melting points were obtained at a 1 °C·min−1 ramp
rate and are uncorrected.
b
“−” symbol indicates the lack of observed proton transfer. MTBD =
7-methyl-1,5,7-triazabicyclo[4.4.0]dec-5-ene.
kcal/mol and below 260.6 kcal/mol, consistent with the
theoretical calculations.
The observance of proton transfer with 6a-cH+ suggested to
us that the absence of proton transfer between 1H+ and 2H+
with BEMP was a kinetic effect due to sterics. Electrostatic
potential (ESP) calculations, which enable a visual representa-
tion of the accessibility of the C2 proton, supported our
hypothesis (Figure 3). In the cases where experimental
bracketing was not possible (i.e., 1H+ and 2H+), strong
shielding of the C2 proton was observed. In contrast, when
smaller aryl substituents were employed, the C2 proton became
more exposed and susceptible to deprotonation even by
sterically bulky bases.
Finally, we calculated the HOMO and LUMO energies for
the DACs and NHCs studied herein. Given the similar PAs yet
dissimilar electrophilicities, we expected that the HOMO
energies would be similar, but not the LUMO energies.22
Indeed, DAC 1 was calculated to display a HOMO that was
only 7 kcal/mol lower than that of 3a; the LUMO of the former
carbene, in contrast, was 54 kcal mol−1 lower than that of the
latter. Likewise, the DAC 2 was calculated to display a HOMO
energy level that was similar to that of NHC 3a (within 6 kcal/
mol) whereas the energy level of the LUMO of the DAC was
51 kcal/mol lower than that of the NHC.
Synthesis of Bis(3-methoxyphenyl)formamidine. A 100 mL
round-bottom flask was charged with m-anisidine (20.0 g, 162 mmol, 2
equiv), triethyl orthoformate (12.0 g, 81 mmol, 1 equiv), five drops of
formic acid, and a stir bar. The flask was equipped with a short-path
distillation head and heated to 110 °C for 2 h, distilling off ethanol.
The temperature was increased to 140 °C for 2 h and then cooled to
room temperature whereupon the reaction mixture crystallized. The
crude product was ground with a glass rod, triturated with pentane,
and filtered through a medium porosity glass frit. Washing with
pentane until the filtrate was colorless followed by drying the collected
solid under reduced pressure afforded the desired compound as a tan
1
solid (19.3 g, 75 mmol, 93%). Mp: 108−109 °C. H NMR (CDCl3,
CONCLUSIONS
400.09 MHz): δ 3.66 (s, 6H), 6.56 (bs, 2H), 6.61−6.66 (m, 4H), 7.18
(t, 3J = 8.0 Hz, 2H), 8.26 (s, 1H), 10.29 (bs, 1H). 13C NMR (CDCl3,
100.50 MHz): δ 55.0, 104.9, 109.3, 111.0, 130.1, 146.6, 150.2, 160.5.
IR (KBr): ν = 3003.9, 2960.5, 2833.7, 1672.9, 1590.1, 1485.6, 1295.2,
1149.6, 1050.2, 1034.9, 779.1 cm−1. HRMS (CI): [M]+ calcd for
C15H16N2O2 256.1212, found 256.1212. Anal. Calcd for C15H16N2O2:
C, 70.29; H, 6.29; N, 10.93. Found: C, 70.26; H, 6.31; N, 11.29.
Synthesis of Bis(4-methoxyphenyl)formamidine. A 100 mL
round-bottom flask was charged with p-anisidine (20.0 g, 162 mmol, 2
equiv), triethyl orthoformate (12.0 g, 81 mmol, 1 equiv), five drops of
formic acid, and a stir bar. The flask was equipped with a short path
distillation head and heated to 110 °C for 2 h, distilling off ethanol.
The temperature was increased to 140 °C for 2 h and then cooled to
room temperature whereupon the reaction mixture crystallized. The
crude product was ground with a glass rod, triturated with pentane,
and filtered through a medium porosity glass frit. Washing with
pentane until the filtrate was colorless followed by drying the collected
solid under reduced pressure afforded the desired compound as a dark
red-brown solid (19.95 g, 78 mmol, 96%). Mp: 118−119 °C. 1H NMR
(CDCl3, 400.09 MHz): δ 3.77 (s, 6H), 6.82 (d, 3J = 9.0 Hz, 4H), 6.95
■
In summary, for the first time, the gas-phase proton affinities of
the diamidocarbenes were assessed both computationally and
experimentally. The inability to bracket the PAs of 1 and 2 was
attributed to steric hindrance at the carbene center by the N-
aryl substituents and supported by electrostatic potential
surface calculations. Moreover, protonated DAC variants with
decreased steric demands, including perfluorophenyl, p-tolyl,
and various anisidyl derivatives, prepared by electrospray of the
formally hydrated carbene precursors in formic acid solution,
were found to undergo deprotonation in the gas phase upon
exposure to appropriate bases. Bracketed PA values were in
good agreement with theory which underscored the ability to
accurately calculate PAs via this method. Except in the case of
the perfluorophenyl derivate 4, which possessed a very low PA
(233.0 kcal/mol), the proton affinities of the DACs studied
herein were similar in value to the PAs of various NHCs
(256.3−261.0 kcal/mol). Collectively, the calculations and the
mass spectrometry data obtained for the carbenes described
herein indicate that the DACs display similar basicities as the
3
(d, J = 9.0 Hz, 4H), 8.07 (s, 1H), 9.93 (bs, 1H). 13C NMR (CDCl3,
100.50 MHz): δ 55.4, 114.5, 120.3, 138.8, 150.0, 155.8. IR (KBr): ν =
10455
dx.doi.org/10.1021/jo401902c | J. Org. Chem. 2013, 78, 10452−10458

The Journal of Organic Chemistry
Article
Figure 3. Calculated electrostatic potential surfaces for various protonated DACs. Values in blue are calculated proton affinities for the
corresponding carbenes (B3LYP/6-31+G(d), kcal/mol).
2907.6, 2832.2, 1669.2, 1518.0, 1441.8, 1295.1, 1249.0, 1205.6, 1028.2,
823.3, 719.3, 526.3 cm−1. HRMS (CI): [M]+ calcd for C15H16N2O2
256.1212, found 256.1212. Anal. Calcd for C15H16N2O2: C, 70.29; H,
6.29; N, 10.93. Found: C, 70.14; H, 6.32; N, 11.09.
(CDCl3, 100.60 MHz): δ 22.8, 23.3, 23.6, 24.3, 25.2, 27.1, 28.7, 29.3,
46.5, 89.6, 123.9, 124.9, 129.5, 132.4, 145.6, 149.2, 171.7. IR (KBr): ν
= 3351.9, 2968.0, 2930.9, 2868.5, 1679.4, 1640.4, 1464.1, 1446.2,
1320.6, 1128.5, 1049.8, 805.7 cm−1. HRMS (CI): [M + H]+ calcd for
C30H43N2O3 479.3274, found 479.3269. Anal. Calcd for C30H42N2O3:
C, 75.28; H, 8.84; N, 5.85. Found: C, 74.87; H, 8.96; N, 5.92.
Synthesis of 4·H2O. A 25 mL Schlenk flask was charged with
bis(perfluorophenyl)formamidine (0.100 g, 0.266 mmol), dichloro-
methane (10 mL), triethylamine (56 μL, 0.399 mmol, 1.5 equiv), and a
stir bar. The mixture was cooled to 0 °C in an ice bath, and then
dimethylmalonyl dichloride (0.047 g, 0.279 mmol, 1.05 equiv) was
added dropwise. The ice bath was then removed, and the reaction was
stirred at ambient temperature for 1.5 h, whereupon the volatiles were
removed under reduced pressure. The solid residue was extracted with
benzene (20 mL) and filtered through a medium frit funnel under
Synthesis of 1·H2O. A 20 mL vial was charged with 2-chloro-1,3-
bis(2,6-diisopropylphenyl)-4,6-diketo-5,5-dimethylpyrimidine (0.150
g, 0.302 mmol) and a stir bar. To this vial open to the atmosphere
was added benzene (5 mL) and stirred at ambient temperature until
the solvent evaporated. Washing with pentane and drying under
reduced pressure afforded the desired product as a white solid (0.127
g, 0.265 mmol, 88%). Mp: 205−206 °C dec. 1H NMR (CDCl3, 400.27
3
MHz): δ 1.13−1.16 (overlapping d, 12H), 1.19 (d, J = 6.8 Hz, 6H),
3
1.29 (d, J = 6.8 Hz, 6H), 1.58 (s, 3H), 1.80 (s, 3H), 2.97−3.07 (d
3
3
overlapping sept, 3H), 3.24 (sept, J = 6.8 Hz, 2H), 5.78 (d, J = 4.1
Hz, 1H), 7.18−7.23 (m, 4H), 7.35 (t, J = 7.7 Hz, 2H). 13C NMR
3
10456
dx.doi.org/10.1021/jo401902c | J. Org. Chem. 2013, 78, 10452−10458

The Journal of Organic Chemistry
Article
nitrogen. The filtrate was concentrated to dryness, redissolved in 1 mL
CHCl3, and stirred open to the atmosphere for 1 h. Removal of
volatiles under reduced pressure, washing with pentane (3 × 1.5 mL),
and drying under reduced pressure afforded the desired product as a
white solid (0.106 g, 0.216 mmol, 81%). Mp: 85−87 °C dec. 1H NMR
(CDCl3, 400.09 MHz): δ 1.58 (s, 3H), 1.82 (s, 3H), 3.65 (bs, 1H),
5.99 (s, 1H). 13C NMR (CDCl3, 100.60 MHz): δ 21.0, 27.2, 48.6, 87.9,
112.3, 138.0, 138.1, 142.1, 144.1, 145.1, 170.9. The aryl 13C signals
(112−145 ppm) were determined via 19F decoupled 13C NMR
spectroscopy. 19F NMR (CDCl3, 376.46 MHz): δ −141.33 (m, 2F),
−145.9 (m, 2F), −150.87 (t, J = 21.8 Hz, 2F), −160.24 to −160.45
(m, 4F). IR (KBr): ν = 3262.9, 1724.2, 1708.5, 1761.2, 1651.7, 1520.7,
1410.5, 1109.8, 994.7, 635.9 cm−1. HRMS (CI): [M + H]+ calcd for
C18H9F10N2O3 491.0453, found 491.0464. Anal. Calcd for
C18H8F10N2O3: C, 44.10; H, 1.64; N, 5.71. Found: C, 43.78; H,
1.75; N, 5.34.
Synthesis of 5·H2O. A 25 mL Schlenk flask was charged with
N,N′-bis(4-methylphenyl)formamidine (0.100 g, 0.446 mmol, 1
equiv), dichloromethane (15 mL), triethylamine (93 μL, 0.668
mmol, 1.5 equiv), and a stir bar. The mixture was cooled to 0 °C in
an ice bath, and then dimethylmalonyl dichloride (0.079 g, 0.468
mmol, 1.05 equiv) was added dropwise. The ice bath was then
removed, and the reaction was allowed to stir at ambient temperature
for 1 h whereupon the volatiles were removed under reduced pressure.
The solid residue was extracted with benzene (25 mL) and filtered
through a medium frit funnel under nitrogen. Concentration of the
filtrate to dryness followed by the addition of 5 mL wet benzene and
stirring open to the atmosphere for 30 min resulted in a precipitate
that was collected over a medium frit funnel. Further purification of
the crude solid by column chromatography (eluent =1:1 v/v hexanes/
ethyl acetate) afforded the desired product as a white solid (0.063 g,
0.186 mmol, 42%). Mp: 133−134 °C. 1H NMR (CDCl3, 399.68
MHz): δ 1.42 (s, 3H), 1.67 (s, 3H), 2.32 (s, 6H), 4.69 (bs, 1H), 5.89
(s, 1H), 7.05 (d, 3J = 8.6 Hz, 4H), 7.13 (d, 3J = 8.2 Hz, 4H). 13C NMR
(CDCl3, 100.50 MHz): δ 21.1, 21.8, 27.2, 47.5, 90.2, 127.3, 130.0,
136.3, 138.0, 172.7. IR (KBr): ν = 3318.3, 2983.1, 2923.1, 1691.8,
1652.5, 1513.9, 1428.2, 1230.6, 1124.2, 1106.8, 1032.1, 753.6, 522.7
cm−1. HRMS (CI): [M − H]+ calcd for C20H21N2O3 337.1552, found
337.1548. Anal. Calcd for C20H22N2O3: C, 70.99; H, 6.55; N, 8.28.
Found: C, 70.96; H, 6.05; N, 8.18.
the reaction was allowed to stir at ambient temperature for 1.5 h,
whereupon the volatiles were removed under reduced pressure. The
solid residue was extracted with benzene (100 mL) and filtered
through a medium frit funnel under nitrogen. Concentration of the
filtrate to dryness followed by the addition of 25 mL of wet benzene,
and stirring open to the atmosphere for 30 min resulted in a
precipitate that was collected over a medium frit funnel. Concentration
of the filtrate and purification of the crude solid by silica gel column
chromatography (eluent =1:1 v/v hexanes/ethyl acetate) afforded the
desired product as a white solid (0.108 g, 0.292 mmol, 15%). Mp: 63−
1
65 °C dec. H NMR (CDCl3, 399.68 MHz): δ 1.46 (s, 3H), 1.69 (s,
3H), 3.75 (s, 6H), 4.64 (bs, 1H), 5.97 (s, 1H), 6.77−6.85 (m, 6H),
7.24 (t, 3J = 8 Hz, 2H). 13C NMR (CDCl3, 100.50 MHz): δ 21.8, 27.3,
47.7, 55.4, 90.1, 113.5, 133.7, 119.6, 130.1, 139.9, 160.3, 172.5. IR
(KBr): ν = 3291.3, 2941.3, 2837.0, 1699.5, 1652.2, 1603.3, 1495.4,
1288.5, 1221.4, 1038.7 cm−1. HRMS (CI): [M]+ calcd for C20H22N2O5
370.1529, found 370.1525. Anal. Calcd for C20H22N2O5: C, 64.85; H,
5.99; N, 7.56. Found: C, 64.64; H, 6.10; N, 7.68.
Synthesis of 6c·H2O. A 25 mL Schlenk flask was charged with
N,N′-bis(4-methoxyphenyl)formamidine (0.100 g, 0.390 mmol, 1
equiv), dichloromethane (15 mL), triethylamine (82 μL, 0.585 mmol,
1.5 equiv), and a stir bar. The mixture was cooled to 0 °C in an ice
bath, and then dimethylmalonyl dichloride (0.069 g, 0.410 mmol, 1.05
equiv) was added dropwise. The ice bath was removed, and the
reaction was allowed to stir at ambient temperature for 1 h whereupon
the volatiles were removed under reduced pressure. The solid residue
was extracted with benzene (25 mL) and filtered through a medium
frit funnel under nitrogen. Concentration of the filtrate to dryness
followed by the addition of 5 mL of wet benzene and stirring open to
the atmosphere for 30 min resulted in a precipitate that was collected
over a medium frit funnel. Further purification of the crude solid by
column chromatography (eluent = 1:2 v/v hexanes/ethyl acetate)
afforded the desired product as a white solid (0.069 g, 0.186 mmol,
48%). Mp: 147−148 °C dec. 1H NMR (CDCl3, 400.09 MHz): δ 1.54
(s, 3H), 1.76 (s, 3H), 3.79 (s, 6H), 3.81 (s, 1H), 6.05 (s, 1H), 6.91 (d,
3J = 8.6 Hz, 4H), 7.20 (d, 3J = 9.0 Hz, 4H). 13C NMR (CDCl3, 100.60
MHz): δ 21.8, 27.5, 47.6, 55.5, 90.5, 114.8, 115.3, 120.9, 129.0, 131.6,
159.3, 172.7. IR (KBr): ν = 3213.1, 2953.9, 2837.2, 1713.1, 1697.6,
1642.6, 1512.3, 1249.6, 1029.6, 833.5 cm−1. HRMS (CI): [M]+ calcd
for C20H22N2O5 370.1529, found 370.1524. Anal. Calcd for
C20H22N2O5: C, 64.85; H, 5.99; N, 7.56. Found: C, 64.47; H, 5.89;
N, 7.83.
Synthesis of 6a·H2O. A 25 mL Schlenk flask was charged with
N,N′-bis(2-methoxyphenyl)formamidine (0.100 g, 0.390 mmol, 1
equiv), dichloromethane (15 mL), triethylamine (82 μL, 0.585 mmol,
1.5 equiv), and a stir bar. The mixture was cooled to 0 °C in an ice
bath, and then dimethylmalonyl dichloride (0.069 g, 0.410 mmol, 1.05
equiv) was added dropwise. The ice bath was then removed, and the
reaction was allowed to stir at ambient temperature for 1 h whereupon
the volatiles were removed under reduced pressure. The solid residue
was extracted with benzene (25 mL) and filtered through a medium
frit funnel under nitrogen. Concentration of the filtrate to dryness
followed by the addition of 5 mL wet benzene and stirring open to the
atmosphere for 30 min resulted in a precipitate that was collected over
a medium frit funnel. Further purification of the crude solid by column
chromatography (eluent =1:3 v/v hexanes/ethyl acetate) afforded the
desired product as an off-white solid (0.074 g, 0.200 mmol, 51%). Mp:
Gas-Phase Experiments and Calculations. Bracketing experi-
ments were conducted using a house-modified quadrupole ion trap
mass spectrometer as previously described.35 Protonated carbene ions
were generated by electrospray ionization (ESI) from a ∼10−4
M
solution using a flow rate of 15−25 μL/min. The analyzed solutions
were prepared by dissolving the conjugate acids of NHCs 3a, 3b, 3c,
and 3d in methanol; dissolving the formally hydrated DACs 1, 4, and 5
in pure formic acid; or dissolving the formally hydrated DACs 2 and 6
in pure formic acid followed by ten-fold dilution using DI water.
The capillary temperature was 150 °C. Neutral reference bases were
added with the helium gas flow. The protonated carbene ions were
allowed to react with neutral reference bases for 0.03−1000 ms. The
unreactive DACs 1 and 2 were allowed to react up to 10000 ms with
BEMP to ensure that proton transfer was not occurring. The
occurrence of proton transfer was regarded as evidence that the
reaction was exothermic (“+” in tables); otherwise, the reaction was
regarded as endothermic (“−” in tables). The typical electrospray
needle voltage was ∼4.5 kV. A total of 10 scans were collected and
averaged.
1
161−162 °C dec. H NMR (CDCl3, 399.68 MHz): δ 1.57 (s, 3H),
1.83 (s, 3H), 3.84 (s, 6H), 4.40 (bs, 1H), 5.92 (s, 1H), 6.97−7.05 (m,
4H), 7.20−7.35 (m, 4H). 13C NMR (CDCl3, 100.65 MHz): δ 21.8,
27.4, 48.2, 56.1 (b), 90.0 (b), 112.3 (b), 121.2 (b), 127.7 (b), 129.9
(b), 154.7, 172.6. IR (KBr): ν = 3244.2, 2943.0, 2840.0, 1689.4,
1645.2, 1504.6, 1431.5, 1272.2, 1128.5, 1044.7, 806.2, 770.8 cm−1.
HRMS (CI): [M]+ calcd for C20H22N2O5 370.1529, found 370.1528.
Anal. Calcd for C20H22N2O5: C, 64.85; H, 5.99; N, 7.56. Found: C,
64.59; H, 5.93; N, 7.75.
All calculations were performed using density functional theory
(B3LYP/6-31+G(d))50−54 as implemented in Gaussian 09.55 All of the
geometries were fully optimized, and frequencies were calculated; no
scaling factor was applied. The optimized structures had no negative
frequencies. The temperature for the calculations was set to 298 K.
GaussView 5.0 was used to generate the electrostatic potential maps of
the protonated carbenes based on their optimized structures in the gas
phase. Density isovalues for the surfaces were set to 0.0004. The color
range for the surfaces was set to −0.19 to +0.19.
Synthesis of 6b·H2O. A 100 mL Schlenk flask was charged with
N,N′-bis(3-methoxyphenyl)formamidine (0.500 g, 1.95 mmol, 1
equiv), dichloromethane (75 mL), triethylamine (0.4 mL 2.92
mmol, 1.5 equiv), and a stir bar. The mixture was cooled to 0 °C in
an ice bath, and then dimethylmalonyl dichloride (0.346 g, 2.05 mmol,
1.05 equiv) was added dropwise. The ice bath was then removed, and
10457
dx.doi.org/10.1021/jo401902c | J. Org. Chem. 2013, 78, 10452−10458

The Journal of Organic Chemistry
Article
(28) Massey, R. S.; Collett, C. J.; Lindsay, A. G.; Smith, A. D.;
O’Donoghue, A. C. J. Am. Chem. Soc. 2012, 134, 20421.
(29) Alder, R. W.; Allen, P. R.; Williams, S. J. J. Chem. Soc., Chem.
Commun. 1995, 1267.
(30) Kim, Y.-J.; Streitwieser, A. J. Am. Chem. Soc. 2002, 124, 5757.
(31) Chu, Y.; Deng, H.; Cheng, J.-P. J. Org. Chem. 2007, 72, 7790.
(32) Gronert, S. Chem. Rev. 2001, 101, 329.
ASSOCIATED CONTENT
■
S
* Supporting Information
Cartesian coordinates for all calculated species and NMR
spectra. This material is available free of charge via the Internet
at http://pubs.acs.org.
(33) Ervin, K. M. Chem. Rev. 2001, 101, 391.
(34) Liu, M.; Yang, I.; Buckley, B.; Lee, J. K. Org. Lett. 2010, 12,
4764.
(35) Liu, M.; Chen, M.; Zhang, S.; Yang, I.; Buckley, B.; Lee, J. K. J.
Phys. Org. Chem. 2011, 24, 929.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: bielawski@cm.utexas.edu.
*E-mail: jee.lee@rutgers.edu.
(36) Preliminary calculations and experiments indicate that the
imidazolinylidene (i.e., saturated) derivatives were less than 1 kcal/mol
more basic than the unsaturated analogues 3a and 3d.
(37) Chen, H.; Justes, D. R.; Cooks, R. G. Org. Lett. 2005, 7, 3949.
(38) These bases include tert-butyliminotri(pyrrolidino)phosphorane
(PA = 266.5 kca/mol), 2,8,9-triisobutyl-2,5,8,9-tetraaza-1-phosphabi-
cyclo[3.3.3]undecane (PA = 268.6 kcal/mol), and tetramethyl(tris(di-
methylamino)phosphoranylidene)phosphorictriamid-Et-imin (PA =
272.3 kcal/mol) (see ref 39).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully thank the NSF (CHE-0954364 to J.K.L.; CHE-
1266323 to C.W.B.), the ACS-PRF, and the National Center
for Supercomputer Applications for support.
(39) Kaljurand, I.; Koppel, I. A.; Kutt, A.; Room, E.-I.; Rodima, T.;
̈
̃
̃
Koppel, I.; Mishima, M.; Leito, I. J. Phys. Chem. A 2007, 111, 1245.
(40) For example, the proton transfer rates with 2,6-disubstituted
tert-butylpyridine are lower than those observed with other substituted
pyridines. See refs 41 and 42.
(41) Hopkins, H. P., Jr.; Jahagirdar, D. V.; Moulik, P. S.; Aue, D. H.;
Webb, H. M.; Davidson, W. R.; Pedley, M. D. J. Am. Chem. Soc. 1984,
106, 4341.
(42) Jasinski, J. M.; Brauman, J. I. J. Am. Chem. Soc. 1980, 102, 2906.
(43) NIST Chemistry WebBook; NIST Standard Reference Database
No. 69; retrieved in 2011; Linstrom, P. J., Mallard, W. G., Eds.;
National Institute of Standards and Technology: Gaithersburg, MD,
http://webbook.nist.gov.
(44) Moerdyk, J. P.; Bielawski, C. W. Nature Chem. 2012, 4, 275.
(45) Moerdyk, J. P.; Bielawski, C. W. J. Am. Chem. Soc. 2012, 134,
6116.
REFERENCES
■
(1) Hirai, K.; Komatsu, K.; Tomioka, H. Chem. Lett. 1994, 23, 503.
(2) Tomioka, H.; Mizutani, K.; Matsumoto, K.; Hirai, K. J. Org.
Chem. 1993, 58, 7128.
(3) Robinson, G. C. Tetrahedron Lett. 1965, 6, 1749.
(4) Sander, W.; Bucher, G.; Wierlacher, S. Chem. Rev. 1993, 93, 1583.
(5) Igau, A.; Grutzmacher, H.; Baceiredo, A.; Bertrand, G. J. Am.
Chem. Soc. 1988, 110, 6463.
(6) Arduengo, A. J., III; Harlow, R. L.; Kline, M. J. Am. Chem. Soc.
1991, 113, 361.
(7) Bourissou, D.; Guerret, O.; Gabbaï, F. P.; Bertrand, G. Chem. Rev.
2000, 100, 39.
̈
(8) Ofele, K.; Tosh, E.; Taubmann, C.; Herrmann, W. A. Chem. Rev.
2009, 109, 3408.
(9) Díez-Gonzal
3612.
́
ez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009, 109,
(46) Schaub, T.; Backes, M.; Radius, U. Organometallics 2006, 25,
4196.
(47) Cole, M. L.; Junk, P. C.; Louis, L. M. J. Chem. Soc., Dalton Trans.
2002, 3906.
(48) Binobaid, A.; Iglesias, M.; Beetstra, D. J.; Kariuki, B.; Dervisi, A.;
Fallis, I. A.; Cavell, K. J. Dalton Trans. 2009, 7099.
(10) Droge, T.; Glorius, F. Angew. Chem., Int. Ed. 2010, 49, 6940.
̈
(11) Dixon, D. A.; Arduengo, A. J., III; Dobbs, K. D.; Khasnis, D. V.
Tetrahedron Lett. 1995, 36, 645.
(12) Clavier, H.; Nolan, S. P. Chem. Commun. 2010, 46, 841.
(13) Vougioukalakis, G. C.; Grubbs, R. H. Chem. Rev. 2010, 110,
1746.
́
(49) Abdou, H. E.; Mohamed, A. A.; Lopez-de-Luzuriaga, J. M.;
Monge, M.; Fackler, J. P., Jr. Inorg. Chem. 2012, 51, 2010.
(50) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(51) Becke, A. D. J. Chem. Phys. 1993, 98, 1372.
(52) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(53) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623.
(14) Valente, C.; Çalimsiz, S.; Hoi, K. H.; Mallik, D.; Sayah, M.;
Organ, M. G. Angew. Chem., Int. Ed. 2012, 51, 3314.
(15) Delaude, L. Eur. J. Inorg. Chem. 2009, 1681.
(16) Hudnall, T. W.; Bielawski, C. W. J. Am. Chem. Soc. 2009, 131,
16039.
(17) Hudnall, T. W.; Moerdyk, J. P.; Bielawski, C. W. Chem.
Commun. 2010, 46, 4288.
(18) Cesar, V.; Lugan, N.; Lavigne, G. Eur. J. Inorg. Chem. 2010, 361.
́
(19) Hobbs, M. G.; Forster, T. D.; Borau-Garcia, J.; Knapp, C. J.;
Tuononen, H. M.; Roesler, R. New J. Chem. 2010, 34, 1295.
(20) Braun, M.; Frank, W.; Reiss, G. J.; Ganter, C. Organometallics
2010, 29, 4418.
(21) Hudnall, T. W.; Tennyson, A. G.; Bielawski, C. W.
Organometallics 2010, 29, 4569.
(22) Hudnall, T. W.; Moorhead, E. J.; Gusev, D. G.; Bielawski, C. W.
J. Org. Chem. 2010, 75, 2763.
(23) Moerdyk, J. P.; Bielawski, C. W. Organometallics 2011, 30, 2278.
(24) Lee, Y.-G.; Moerdyk, J. P.; Bielawski, C. W. J. Phys. Org. Chem.
2012, 25, 1027.
(54) Kohn, W.; Becke, A. D.; Parr, R. G. J. Phys. Chem. 1996, 100,
12974.
(55) Frisch, M. J.; et al. Gaussian 09; Gaussian, Inc.: Wallingford, CT,
2009.
NOTE ADDED AFTER ASAP PUBLICATION
■
Due to a production error, the last sentence in the Results and
Discussion section was truncated in the version published on
October 3, 2013. The sentence is completed in the version
reposted on October 4, 2013.
(25) Washabaugh, M. W.; Jencks, W. P. Biochemistry 1988, 27, 5044.
(26) Amyes, T. L.; Diver, S. T.; Richard, J. P.; Rivas, F. M.; Toth, K. J.
Am. Chem. Soc. 2004, 126, 4366.
(27) Higgins, E. M.; Sherwood, J. A.; Lindsay, A. G.; Armstrong, J.;
Massey, R. S.; Alder, R. W.; O’Donoghue, A. C. Chem. Commun. 2011,
47, 1559.
10458
dx.doi.org/10.1021/jo401902c | J. Org. Chem. 2013, 78, 10452−10458