Catalytic Halogen Bond Activation in the Benzylic C-H Bond Iodination with Iodohydantoins

Article abstract of DOI:10.1021/acs.orglett.7b03034

This letter presents the side-chain iodination of electron-deficient benzylic hydrocarbons at rt using N-hydroxyphthalimide (NHPI) as radical initiator and 1,3-diiodo-5,5-dimethylhydantoin and 3-iodo-1,5,5-trimethylhydantoin (3-ITMH) as iodine source. Addition of a carboxylic acid increased the reactivity due to complex formation with and activation of 3-ITMH by proton transfer and halogen bond formation. No S_EAr reactions were observed under the employed reaction conditions. Our method enables convenient product isolation and gives 50-72% yields of isolated products.

Full text of DOI:10.1021/acs.orglett.7b03034

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX-XXX
Catalytic Halogen Bond Activation in the Benzylic C−H Bond
Iodination with Iodohydantoins
Sascha H. Combe, Abolfazl Hosseini, Lijuan Song, Heike Hausmann, and Peter R. Schreiner*
Institute of Organic Chemistry, Justus-Liebig University, Heinrich-Buff-Ring 17, 35392 Giessen, Germany
S
* Supporting Information
ABSTRACT: This letter presents the side-chain iodination of
electron-deficient benzylic hydrocarbons at rt using N-hydroxyph-
thalimide (NHPI) as radical initiator and 1,3-diiodo-5,5-dimethyl-
hydantoin and 3-iodo-1,5,5-trimethylhydantoin (3-ITMH) as iodine
source. Addition of a carboxylic acid increased the reactivity due to
complex formation with and activation of 3-ITMH by proton
transfer and halogen bond formation. No SEAr reactions were
observed under the employed reaction conditions. Our method
enables convenient product isolation and gives 50−72% yields of
isolated products.
arguably the best protocol for benzylic substrates bearing an
ethods for the iodination of unactivated alkanes are
Mscarce because the initiation and propagation steps with electron-donating group (EDG), but NaN3 is toxic as well as
explosive. Earlier, we described a phase-transfer single electron
(SET) approach with crushed NaOH and HCI3, generating
chain-carrying •CI3 radicals.1a With this method, aliphatic
substrates are easily iodinated in good yields and with very high
regioselectivities, which results from the sterically demanding
•CI3 radical. Very recently, Gandelman and Schreiner
presented the iodination of aliphatic and benzylic substrates
with 1-iodo-3,5,5-trimethylhydantoin (1-ITMH) (1) at 100 °C
neat (alkanes) or in benzene in combination with a 3 W LED
lamp (Scheme 1c).7 The benzylic iodoalkanes were thereby
transformed in situ to esters as their isolation was low yielding.
Commonly, benzyl iodides are synthesized starting from benzyl
alcohols8 or benzyl chlorides9 because the direct iodination of
unactivated alkanes, especially isolation of the pure compound,
remains a challenging task (Scheme 1d). Hence, there is still a
need for nontoxic, safe, and catalytic methods that enable
practical product isolation.
Orazi et al. first used the currently commercially available
iodination agent 1,3-diiodo-5,5-dimethylhydantoin (DIH, 2) in
the electrophilic core iodination of arenes.10 When the authors
treated toluene with 2 in combination with benzoyl peroxide
(BPO), no benzyl iodide was observed, and they concluded
that 2 only serves as a source for electrophilic “I+”.10 Indeed,
there are only a few examples where 2 can also act as an iodine
radical source.11 The N−I bonds in 1 (BDE = 43.5 kcal
mol−1),7 2 (amide 38.1 kcal mol−1 and imide 48.7 kcal
mol−1),12 and 3-iodo-1,5,5-trimethylhy-dantoin (3-ITMH, 3)
(54.6 kcal mol−1)7 are quite low compared to that of N-
iodosuccinimide (56.8 kcal mol−1)7 and are therefore ideal
candidates for the radical iodination of hydrocarbon C−H
elemental iodine are endothermic and not counterbalanced by
the heat released from iodoalkane formation owing to the
relatively C−I low BDEs.1 Furthermore, the use of I2 affords
hydrogen iodide that may immediately reduce the iodoalkane
to the corresponding alkane.2 Photochemically initiated
iodinations suffer from the cleavage of the C(sp3)−I bond
due to n- to σ*-excitation at ∼250 nm and are therefore
unfavorable.3 This results in iodoalkanes being relatively
expensive yet highly valuable reactants in organic synthesis.
Only in 1968 did Tanner and co-workers succeed in
synthesizing iodoalkanes from unactivated alkanes by the use
of tert-butyl hypoiodite.4 Most existing methods use tert-butyl
hypoiodite as an intermediate to generate the strongly
electrophilic tert-butoxy radical or the corresponding iodine-
(III) radical (Scheme 1a).4,5 Unfortunately, tert-butyl hypo-
iodite must be prepared in situ due to its instability.4 Barluenga
et al. used PhI(OAc)2, I2, and catalytic amounts of trimethylsilyl
azide (TMSN3) at 60 °C to form PhIOAc radicals that enable
homolytic C−H bond cleavage (Scheme 1b).6 In the next step,
I2 reacts with the carbon radical to form the iodoalkane.
Barluenga and co-workers described a second variant consisting
of H2O2, 3 equiv of NaN3, and I2 at 0−40 °C. To date, this is
Scheme 1. Iodination Methods a−c; Common Synthesis d
Received: September 28, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b03034
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Initial Evaluation of Reaction Conditions for the
Side-Chain Iodination of Toluene
Table 2. HOMO−SOMO Energy Gap (ΔE) at B3LYP/6-
31G(d,p)
5
6
PhCHO
a
a
R
-H
-Cl
-CO2Me
1.45
-CF3
1.40
-CN
1.21
-NO2
0.99
entry t (h) (equiv)
cat. (equiv)
(%)
(%)
b
ΔE (eV)
1.92
1.73
1
2
3
4
5
6
7
8
9
16
16
1.0
1.0
1.7
1.7
1.7
1.7
1.7
1.7
3.4
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
6.7
Cu(OAc)2·H2O (0.03)
Cu(OAc)2·H2O (0.03)
11
19
18
27
13
33
38
39
60
69
8
3
3
c
Scheme 3. Proposed Reaction Mechanism for the Side-
Chain Iodination of Toluene with Computations at the
B3LYP-D3/6-31G(d,p) Level of Theory (ΔG298K in kcal
mol−1)
16
−
−
−
−
−
−
−
−
2
d
e
f
16
3
16
<1
3
16
g
h
16
1
16
1
16
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
16
1
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
−
−
traces
2
PhCO2H (0.18)
PhCO2H (0.36)
100% AcOH (0.36)
Ac2O (0.18)
diphenic acid (0.36)
p-NO2C6H4CO2H (0.36)
o-NO2C6H4CO2H (0.36)
Cl2HCO2H (0.36)
CSA (0.36)
p-TsOH (0.36)
TFA (0.36)
MsOH (0.36)
Ph3PS (0.36)
36
63
62
13
65
15
69
47
12
12
21
−
1
1
1
<1
<1
<1
1
1
iodinations because triiodoisocyanuric acid (TICA) is not
commercially available and its synthesis is cumbersome.14
Compared to TCCA, TICA is also extremely moisture
sensitive.14b We have chosen toluene (5) as our model
substrate for our new iodination protocol. N-iodosuccinimide
(NIS) afforded benzyl iodide (6) in low yield (Table 1, entry
1). However, 2 afforded approximately twice as much product
as that of NIS (entry 2). In our TCCA chlorination, a metal
catalyst was crucial to perform the reaction. Gratifyingly, no
metal catalyst is required with 2, and the yields were higher
compared to those with NIS (entry 6). Compound 2 is a strong
oxidant that is able to generate phthalimide-N-oxyl (PINO, 7)
radical from 4. High dilution provides higher yields (entries 3,
4, and 6). A loading of 4 of ∼17 mol % is optimal for a smooth
reaction (entry 6); lower loadings down to 8 mol % decrease
the yields (entry 5), and higher loadings provide no additional
benefit (entries 7 and 8). High substrate loadings also gave
higher yields (entries 9 and 10). Using CHCl3, 1,2-dichloro-
ethane or CH3CN gave inferior results; the latter mainly
promoted SEAr iodination. To activate the N−I bonds of 2, we
screened several carboxylic and sulfonic acids (entries 12−
23).15 In general, carboxylic acids afforded much higher yields
than the sulfonic acids, which can be correlated to their pKa
values. Strong acids such as methanesulfonic acid led to SEAr
<1
<1
−
−
a
Yield determined by 1H NMR with dimethyl terephthalate (DMT) as
internal standard. NIS (0.625 mmol, 1 equiv) used instead of 2.
Using 1 mL of CH2Cl2. Using 1.5 mL of CH2Cl2. Using 0.08 equiv
of 4. Average H NMR yield of two runs. Using 0.25 equiv of 4.
b
c
d
e
f
g
1
h
Using 0.33 equiv of 4.
Scheme 2. Substrate Scope for the Side-Chain Iodination of
Toluene Derivatives with Yields of Isolated Product Given
reactions. These are known to produce “I+” and I3 , which give
+
rise to core substitution.16 With the best reaction conditions in
hand (entry 18), we explored the substrate scope of the
reaction (Scheme 2). Arenes bearing electron-withdrawing
(EWGs) or weakly electron-donating groups (EDGs) gave
yields of 35−69%. Yields higher than 69% with 2 could not be
achieved, which may be an indication that not all iodine atoms
in 2 are active and transferable. To support our assumption, we
performed NMR studies with compounds 1−3 (Scheme S1).
Compound 1 gave the lowest yield (45%), and 2 and 3 behaved
a
b
Using 2.5 mmol substrate. Without acid.
bonds. Recently, we presented the side-chain chlorination of
aliphatic and benzylic C−H bonds with trichloroisocyanuric
acid (TCCA) and N-hydroxyphthalimide (NHPI, 4) as radical
initiator.13 This method is not directly transferable to
B
DOI: 10.1021/acs.orglett.7b03034
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 1. NMR studies of the complexation of 3-ITMH (3) and acid in CD2Cl2 (62 mM) at 25 °C.
similarly (71 and 81%, respectively). These results imply that
the iodine next to the geminal dimethyl-group in 2 is largely
inactive and that the corresponding N-centered radical is less
reactive for chain propagation.
On the basis of the results presented here, DFT
computations, and literature reports, we propose the
mechanism shown in Scheme 3. With the use of 2 or 3, no
metal catalyst was needed to start the reaction. Hence, 3 must
be able to oxidize 4 to 7. Compound 7 serves in the radical
side-chain chlorination only as chain initiator, and higher
loadings of 4 did not show significant effects; an additional
portion of 4 after 1 h had no beneficial effect, thereby
supporting our suggestion. Furthermore, GC-MS analysis
revealed the formation of phthalimide. The addition of acid
increased the reaction rate considerably (Table 1, entry 18).
For whatever reason, p-nitrobenzoic acid did not show the
same catalytic efficiency as o-nitrobenzoic acid (Table 1, entries
17 and 18). The carboxylic acid does not only serve as a proton
source (general acid catalysis) (Figure 1 and Table S3) but is
likely to additionally form a halogen bond20 (for complex 19,
also see ToC); both interactions polarize and weaken the N−I
bond.16b,21 DFT reveals a bonding angle of ∼152°
(N−I···ON), which is slightly below typical bonding angles
in association with I (160−178°)20b and a short bond distance
of 3.06 Å (Σ r(OIVdW): 3.50 Å).22 The dual binding mode is
only seen when o-nitrobenzoic acid was used (Figure 1, II vs
III). The N-CH3 group of 3 is shifted downfield more strongly
as compared to the geminal dimethyl group (Figure 1, III).
This is likely due to the fact that the carbamide group allows for
better charge delocalization as compared to that of the amide
group. A DFT study (B3LYP/6-311G(d)) of 2 showed that
protonation at the carbamide CO is ∼6 kcal mol−1 more
favorable than at the amide CO.12 In line with this
complexation, the aromatic protons are shifted upfield, gaining
electron density from 3 (Figure 1, III vs IV). The 13C NMR
spectrum shows that 3 donates electron density to the carbonyl
group of the acid (Figure 1, III vs IV) resulting in a strong
upfield shift. Complex formation is also supported by 1H DOSY
NMR (Figure S1 and Table S4). Both molecular components
of the mixture are associated with the same diffusion coefficient
value (which is lower than the values obtained for the free
components in separate experiments). The self-diffusion
coefficients for the free molecular components are at least
Therefore, we repeated all of the reactions with method B
(Scheme 2), substituting 2 with 3; we also increased the
substrate loading, which had the highest influence on the yields
as evident from the data analysis (Tables S1 and S2). In
general, 3 afforded similar results, giving yields in the range of
50−72%. However, 3 gave higher yields compared to 2
especially for substrates bearing strong deactivating groups.
As electron-rich as well as electron-deficient arenes17
undergo SEAr reactions in the presence of acid, our protocol
is quite practical. However, arenes bearing modest to strong
electron-donating groups form the corresponding nitroesters.
Usually, EWG-substituted arenes undergo SN2 reactions much
faster than EDG-substituted arenes because EWGs lower the
SN2 reaction barrier.18 Ethylbenzene gave the nitroester in 24%
yield; p-bromoethylbenzene gave traces of the desired product,
whereas p-cyanoethylbenzene delivered 59% of corresponding
benzyl iodide 15.
The underlying reactivity is not merely a question of BDEs
(ArCH2-H: p-methyl anisole BDE = 87 kcal mol−1; toluene =
88 kcal mol−1)19 or radical stability (14 vs 15) but rather a
question of frontier orbital interaction (Table 2). The results
show that primarily arenes carrying an EWG undergo a smooth
reaction (low-lying SOMO). No SEAr products were detected
for EWG-containing arenes in preparative mass-balanced
reactions; products and starting material (substrate) were
recovered with a mass loss of 7−10%. The intermediate
electrophilic radicals have to react with a nucleophilic iodine
intermediate (Table 2 and Table S6). The three methyl groups
of 3 may play an important role because they can donate
electron density toward the carbonyl groups. Our computa-
tional studies for the side-chain chlorination with TCCA
showed that the reaction proceeds preferentially via a carbonyl
radical13 that abstracts a H-radical from the substrate forming
an iminol, which tautomerizes to the amide.
C
DOI: 10.1021/acs.orglett.7b03034
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(c) Barluenga, J.; Gonzal
́
ez-Bobes, F.; Gonzal
́
ez, J. M. Angew. Chem.,
half their values in the complex (Table S4). The acid
experiences the greatest change in the diffusion coefficient
when complexed (a factor of 2), a fact that agrees with its
participation in the complex. The measurements were carried
out in dilute solution in CD2Cl2, where viscosity variations due
to small sample composition changes are negligible. Thus, the
significant decrease in diffusion for o-nitrobenzoic acid when 3
is present provides strong evidence for the existence of the
complex in solution.
Int. Ed. 2002, 41, 2556. (d) Montoro, R.; Wirth, T. Org. Lett. 2003, 5,
4729. (e) Montoro, R.; Wirth, T. Synthesis 2005, 1473. (f) Ling, X.-G.;
Xiong, Y.; Zhang, S.-T.; Huang, R.-F.; Zhang, X.-H. Chin. Chem. Lett.
2013, 24, 45.
(6) Barluenga, J.; Campos-Gom
́ ́
ez, E.; Rodríguez, D.; Gonzalez-
Bobes, F.; Gonzalez, J. M. Angew. Chem., Int. Ed. 2005, 44, 5851.
́
(7) Artaryan, A.; Mardyukov, A.; Kulbitski, K.; Avigdori, I.;
Nisnevich, G. A.; Schreiner, P. R.; Gandelman, M. J. Org. Chem.
2017, 82, 7093.
Furthermore, I2 acts as a mild Lewis acid catalyst (halogen
bond donor)23 but is not able to oxidize 4 to 724 (Table S5).
The monoacid complex (19) is slightly preferred over the
diacid complex by 0.3 kcal mol−1 (Schemes S4 and S5). There
is no halogen bond in diacid complex 2a-E (Scheme S5). An
acid amount of 2.2 equiv compared to 0.36 equiv gave a rate
acceleration of only ∼2 even at higher dilution. This result also
supports the formation of a monoacid complex. Structure 19
consumes one benzyl radical (20) and forms benzyl iodide (6)
as well as a new radical 21 (ΔG298K = 1.3 kcal mol−1), which is
able to carry the radical chain (cycle A). The latter reacts
further to trimethylhydantoin (22) forming a strong amide
(8) (a) Khazdooz, L.; Zarei, A.; Aghaei, H.; Azizi, G.; Gheisari, M. M.
Tetrahedron Lett. 2016, 57, 168. (b) Bandgar, B. P.; Bettigeri, S. V.
Monatsh. Chem. 2004, 135, 1251.
(9) (a) Namavari, M.; Satyamurthy, N.; Phelps, M. E.; Barrio, J. R.
Tetrahedron Lett. 1990, 31, 4973. (b) Naeimi, H.; Nazifi, Z. S. Russ.
Chem. Bull. 2015, 64, 1814.
(10) Orazi, O. O.; Corral, A.; Bertorello, H. E. J. Org. Chem. 1965, 30,
1101.
(11) (a) Baba, H.; Togo, H. Tetrahedron Lett. 2010, 51, 2063.
(b) Krabbe, S. W.; Do, D. T.; Johnson, J. S. Org. Lett. 2012, 14, 5932.
(c) Bloodworth, A. J.; Curtis, R. J. J. Chem. Soc., Chem. Commun. 1989,
173. (d) Kulbitski, K.; Nisnevich, G.; Gandelman, M. Adv. Synth. Catal.
2011, 353, 1438.
(12) Filimonov, V. D.; Krasnokutskaya, E. A.; Poleshchuk, O. K.;
Lesina, Y. A.; Chaikovskii, V. K. Russ. Chem. Bull. 2006, 55, 1328.
(13) Combe, S. H.; Hosseini, A.; Parra, A.; Schreiner, P. R. J. Org.
Chem. 2017, 82, 2407.
(14) (a) Ribeiro, R. d. S.; Esteves, P. M.; de Mattos, M. C. S.
Tetrahedron Lett. 2007, 48, 8747. (b) Gottardi, W. Monatsh. Chem.
1970, 101, 655.
(15) Jakab, G.; Hosseini, A.; Hausmann, H.; Schreiner, P. R. Synthesis
2013, 45, 1635.
bond, which is the driving force of the reaction (ΔG298K
=
−20.0 kcal mol−1).
In summary, we present a method for the direct C−H bond
iodination of electron-deficient arenes. Iodinating agent 3 can
be activated by a carboxylic acid, polarizing the N−I bond
through simultaneous hydrogen and halogen bonding.
Compared to other methods, our protocol enables convenient
product isolation without postmodification.
(16) (a) Chaikovskii, V. K.; Filimonov, V. D.; Funk, A. A.;
Skorokhodov, V. I.; Ogorodnikov, V. D. Russ. J. Org. Chem. 2007,
43, 1291. (b) Gottardi, W. Monatsh. Chem. 1975, 106, 1203.
(17) (a) Groweiss, A. Org. Process Res. Dev. 2000, 4, 30. (b) de
Almeida, L. S.; de Mattos, M. C. S.; Esteves, P. M. Synlett 2013, 24,
603.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b03034.
■
S
(18) Galabov, B.; Nikolova, V.; Wilke, J. J.; Schaefer, H. F.; Allen, W.
D. J. Am. Chem. Soc. 2008, 130, 9887.
Experimental section, spectra of all compounds,
computations and procedures, Cartesian coordinates,
and absolute energies of all optimized structures (PDF)
(19) Nam, P.-C.; Nguyen, M. T.; Chandra, A. K. J. Phys. Chem. A
2005, 109, 10342.
(20) (a) Lommerse, J. P. M.; Stone, A. J.; Taylor, R.; Allen, F. H. J.
Am. Chem. Soc. 1996, 118, 3108. (b) Cavallo, G.; Metrangolo, P.;
Milani, R.; Pilati, T.; Priimagi, A.; Resnati, G.; Terraneo, G. Chem. Rev.
2016, 116, 2478. (c) Nicolas, I.; Barriere, F.; Jeannin, O.; Fourmigue,
̀ ́
M. Cryst. Growth Des. 2016, 16, 2963. (d) Sun, X.; Wang, W.; Li, Y.;
Ma, J.; Yu, S. Org. Lett. 2016, 18, 4638.
(21) Prakash, G. K. S.; Mathew, T.; Hoole, D.; Esteves, P. M.; Wang,
Q.; Rasul, G.; Olah, G. A. J. Am. Chem. Soc. 2004, 126, 15770.
(22) Rowland, R. S.; Taylor, R. J. Phys. Chem. 1996, 100, 7384.
(23) Laurence, C.; Graton, J.; Berthelot, M.; El Ghomari, M. J. Chem.
- Eur. J. 2011, 17, 10431.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: prs@uni-giessen.de
ORCID
Peter R. Schreiner: 0000-0002-3608-5515
Notes
The authors declare no competing financial interest.
(24) Masui, M.; Ueshima, T.; Ozaki, S. J. Chem. Soc., Chem. Commun.
1983, 479.
ACKNOWLEDGMENTS
This work was supported by the Justus-Liebig University.
■
REFERENCES
■
(1) (a) Schreiner, P. R.; Lauenstein, O.; Butova, E. D.; Fokin, A. A.
Angew. Chem., Int. Ed. 1999, 38, 2786. (b) Hartung, J. Science of
Synthesis 2007, 35, 541.
(2) Niclas, H. J.; Zeuner, F.; Grundemann, E.; Zolch, L. J. Prakt.
̈
̈
Chem. 1987, 329, 400.
(3) Ross, P. L.; Johnston, M. V. J. Phys. Chem. 1995, 99, 4078.
(4) Tanner, D. D.; Gidley, G. C. J. Am. Chem. Soc. 1968, 90, 808.
(5) (a) Tanner, D. D.; Blackburn, E. V.; Reed, D. W.; Setiloane, B. P.
J. Org. Chem. 1980, 45, 5183. (b) Tanner, D. D.; Gidley, G. C.; Das,
N.; Rowe, J. E.; Potter, A. J. Am. Chem. Soc. 1984, 106, 5261.
D
DOI: 10.1021/acs.orglett.7b03034
Org. Lett. XXXX, XXX, XXX−XXX