Angewandte

Research Articles Chemie

www.angewandte.org

How to cite: Angew. Chem. Int. Ed. 2023, 62, e202307372

Halogen Bonding doi.org/10.1002/anie.202307372

N X···O N Halogen Bonds in Complexes of N-Haloimides and

Pyridine-N-oxides: A Large Data Set Study
Rakesh Puttreddy,* J. Mikko Rautiainen, Shilin Yu, and Kari Rissanen*

structural and functional diversity.[4] Due to the paramount

Abstract: N X··· O N + halogen-bonded systems formed significance of NCIs, extensive research has been conducted
by 27 pyridine N-oxides (PyNOs) as halogen-bond (XB) to elucidate and quantify various NCI motifs, which are
acceptors and two N-halosuccinimides, two N-haloph- broadly classified into four categories;[5] van der Waals
thalimides, and two N-halosaccharins as XB donors are forces, dipole-dipole interactions, electrostatic interactions,
studied in silico, in solution, and in the solid state. This and ionic bonds (Figure 1). The strengths of these NCIs vary
large set of data (132 DFT optimized structures, 75 greatly, spanning over a wide range. At the one end of the
crystal structures, and 168 1H NMR titrations) provides spectrum, there are weak NCIs such as the hydrophobic
a unique view to structural and bonding properties. In effect (< 5 kJ mol 1), while the opposite end of the spectrum
the computational part, a simple electrostatic model are the electrostatic interactions, namely, ionic bonds (>
(SiElMo) for predicting XB energies using only the 500 kJ mol 1). Those in the middle of the interaction
properties of halogen donors and oxygen acceptors is spectrum, such as hydrogen and halogen bonds (XB) (~ 0.5-
developed. The SiElMo energies are in perfect accord > 200 kJ mol 1), have received broad interest because they
with energies calculated from XB complexes optimized are ubiquitous in nature and are, together with metal
with two high-level DFT approaches. Data from in silico coordination, the key tools in the creation of complex
bond energies and single-crystal X-ray structures corre- supramolecular systems.[6]
late; however, data from solution do not. The poly- Both HBs and XBs are reversible, and often multiple
dentate bonding characteristic of the PyNOs’ oxygen HBs and XBs are required to provide structural integrity
atom in solution, as revealed by solid-state structures, is and stability to supramolecular materials.[6] The seminal
attributed to the lack of correlation between DFT/solid- work of Margaret Etter et al.[7] has shown that strong HBs
state and solution data. XB strength is only slightly are a reliable design tool for assembling molecules into high
affected by the PyNO oxygen properties [(atomic charge fidelity supermolecules and supramolecular structures. Since
(Q), ionization energy (Is,min) and local negative minima these early studies, a supramolecular approach has been
(Vs,min)], as the σ-hole (Vs,max) of the donor halogen is extensively used to investigate the breadth, scope, and the
the key determinant leading to the sequence N-halosac- importance of HBs in co-crystals or molecular
charin > N-halosuccinimide > N-halophthalimide on the supramolecular complexes. According to Jeffrey[8] HBs can
XB strength. be broadly categorised into weak, moderate and strong
based on the nature of the donor-acceptor pairs[9] and
interaction energies; (1) weak HBs of < 16 kJ mol 1, gener-
ally formed between soft acids and soft bases (e.g., C H···π,
Introduction C H···S), have nearly the same strength as dipole-dipole
interactions. (2) Moderate HBs (16–67 kJ mol 1), which are
Non-covalent interactions (NCIs) play a crucial role in most common and abundant, are formed by combining one
preserving the structural integrity of numerous chemical[1] hard acid and one soft base component or vice versa (e.g.,
and biological[2,3] systems. For example, nature creates a N H···O, N H···halogen). (3) Strong HBs (67- >
diverse range of proteins using the same set of 20 amino
acids that differ only in their side chains. These side chains
induce and support a wide range of NCIs including hydro-
gen bonds (HBs) and ionic bonds, manifesting large

[*] R. Puttreddy, J. M. Rautiainen, S. Yu, K. Rissanen

University of Jyvaskyla, Department of Chemistry
P.O. BOX 35, 40014 Jyväskylä (Finland)
E-mail: [email protected]
[email protected]
© 2023 The Authors. Angewandte Chemie International Edition
published by Wiley-VCH GmbH. This is an open access article under
the terms of the Creative Commons Attribution License, which
permits use, distribution and reproduction in any medium, provided
the original work is properly cited. Figure 1. General hierarchy of chemical bonding.

Angew. Chem. Int. Ed. 2023, 62, e202307372 (1 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

167 kJ mol 1) formed between hard acids and hard bases

(e.g., [F···H···F] +, [N···H···N] +) are as strong as half the
strengths of covalent bonds. The HB and most recently also
XB toolbox has now matured to a point where structures of
any strength can be modelled and attributed to rationalize
chemical transformations ranging from simple organic
reactions[10] to polymers[11] as well as to explain the func-
tional properties of materials.[12]
Halogen bonding[13] is the newest member of the mid- Figure 2. Depiction of electron-acceptor character of the (a) pyridinic N
atom, (b) carbonyl and ether containing O systems, and (c) pyridine N-
spectrum interactions family and a very close relative of
oxide O atom.
HB.[14,15] Although XB and HB are comparable in many
aspects, the differences between the two has boosted the XB
research. There are at least three obvious differences
between XB and HB. First, the polarization and thus XB has been demonstrated in protein-drug complexes[54,55] and
strength of the halogen follows the order I > Br > Cl @ F, on crystal engineering.[56–64] These X···O interactions can be
a given system D X···A (D = any atom that can polarize the classified as weak and moderate XBs with interaction
X; X = electron-deficient I, Br or Cl; A = electron-rich atom, energies 0.5–20 kJ mol 1 and ~ 20–60 kJ mol 1, respectively.
an XB acceptor).[16] Second, modulating the electron-with- Compared to XB research on X···N, X···O systems that
drawing nature of D the electron-deficient area (so called σ- utilize pyridine N-oxides (PyNOs), the oxidised products of
hole[17,18]) can be tuned.[19] Finally, the halogen participating the parent pyridines, as the XB acceptors in C I··· O N + or
in the XB creates a highly directional halogen bond due to N I··· O N + systems is surprisingly scarce.[65–76]
the electron-deficient area (σ-hole) on the extension of the In our earlier contributions, we have investigated
D X bond.[17] The σ-hole imposes constraints to bonding moderate strength C I··· O N + XBs involving 1,n-perfluor-
and linearity of the geometries of XB interactions, allowing oalkyliodoalkanes and PyNOs,[72] as well as strong
design and construction of specific molecular structures and N I··· O N + XBs generated by N-halosaccharins and N-
materials for targeted applications.[20,21] The X···A halogen oxides (Figure 3).[76] The C I··· O N + systems[52] have inter-
bonding motif manifests a wide range of interaction action energies ranging from 31.9 to 46.5 kJ mol 1,[69] while
energies, from weak (~ 0.5–20 kJ mol 1) to moderate (~ 20– the N I··· O N + have values ranging from 47.5 to
60 kJ mol 1) and strong (~ > 60 kJ mol 1) to be utilized in a 120.3 kJ mol 1. The strongest N I··· O N + XBs
multitude of applications.[20,21] The development of the 1 [76]
(120.3 kJ mol ) approach in strength the extremely strong
halogen σ-hole theory and its directionality have signifi- [N X N] + XBs, being nearly equal to the values observed
cantly broadened our understanding of non-covalent inter- (> 160 kJ mol 1)[36] for the 3-center-4-electron [N X N] +
actions, moving beyond the recently discovered chalcogen (X = Br,I) halogen bonds in Barluenga-type halogen(I)
bonding[22] to a whole new palette of previously unforeseen complexes. In related systems N-haloimide/halosulfonimide
interactions like aerogen,[23] spodium,[24] pnictogen,[25] XB donors have been utilized by Fourmigué et al.[77,78]
tetrel,[25] and triel[26] bonds. Although applications are less Cinčić et al.[79–81] and others[82] to make complexes with
obvious, these late-discovery interactions have enormous pyridines as XB acceptors.
technical potential in areas like catalysis, anion recognition, Our attempts[74] to tune the N I··· O N + XB strengths
and medicinal chemistry.[25,27–30] using a broad selection of aromatic N-oxides, showed that
The weak and moderate XBs of C I···N type, formed by the influence of changing N-oxide to the XB strengths is
iodoperfluoroalkane/arenes and N-heterocycles, have been limited compared to the choice of XB donors in
used in functional materials[31] owing to their reversible C I··· O N + or N I··· O N + systems. For example, titration
nature, whereas the strong [N I N] + 3-center-4-electron study of 2-methyl-, 4-phenyl-, 2-methyl-4-nitro-, and parent
halogen bonds stabilize the highly reactive halenium (I +) ion pyridine N-oxide with N-iodosuccinimide (NIS) gave associ-
allowing iodine(I) complexes, such [bis(pyridine)iodine ation constants (KXB) of 779, 325, 17, and 660 M 1,
(I)]BF4 (Barluenga’s reagent) to be used as respectively, while the same N-oxides with stronger XB-
halogenating[32–34] and oxidising agents[35] in organic syn- donor N-iodosaccharin (NISac) gave KXB values of 16338,
thesis. Over the last decade, the + I N XBs in [N I N] + has 14200, 543, and 3121 M 1. These KXB values accurately
received much attention[36] and they are now being utilized follow the value of the σ-hole of NIS and NISac, respec-
for constructing capsules,[37,38] helicates[39] and metal–organic tively. Due to this we became intrigued to investigate to
frameworks (MOFs)[40] unrestrained homoleptic[41,42] and which extent the N I··· O N + XBs can be tuned by choice
heteroleptic[43–45] and hierarchical,[46] and nucleophilic inter- of XB donor, namely the N-haloimides. In this study, we
actions of iodine(I) complexes,[47–49] as well as stable non- prepared a large family of XBs complexes from six N-
chiral[50–52] and chiral carbonyl hypoiodites.[53] Compared to haloimides and 27 PyNOs (Figure 3) to examine the
the much more explored C I···N XB systems,[44] oxygen structures and strengths of N I··· O N + XBs. We chose
atom as an XB acceptor has not attracted so much intention PyNOs with methyl-, methoxy- and phenyl-substituents in
due to its lower nucleophilicity and inherent polydentate ortho-, meta- and para-positions at the pyridine-N-oxide.
nature and electron ’’push-pull’’ property (Figure 2). How- The present work is an expansion of our earlier work[76] on
ever, the significance of X···O(carbonyl/ether) interactions similar system and employs N-haloimides because, the

Angew. Chem. Int. Ed. 2023, 62, e202307372 (2 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

DESiElMo ¼ 0:95ðV S,max � QÞ 1760 � 1=I S,min þ 220

The above study also revealed that 22 of the DFT

optimized XB structures manifest, in addition to the
principal halogen bonding, secondary intramolecular inter-
actions such as C H···O=C and C H···π that further stabilize
the complexes and cause biased XB interaction energies
calculated using DFT methods ΔEDFT, leading to their
exclusion from the SiElMo model. The strong correlation
between the Vs,max value describing the extent of halogen σ-
hole and the strength of the XB, namely the donor halogen
with larger Vs,max value forming stronger XBs has been
observed in halo-perfluoroalkanes/aromatics and N/O/S-
heterocycles. For the N-haloimide XB donors of the current
study, the calculated Vs,max values (Figure 4a) unambiguously
confirm the expected XB donor strength order of: NISac >
NIS > NIP and NBSac > NBS > NBP. The Vs,max for NISac
(+ 186 kJ mol 1) and NBSac (+ 142 kJ mol 1) are higher than
those of commonly used perfluorohalobenzene donors, e.g.,

Figure 3. The chemical structures of the used components. The

aromatic N-oxides (1–27) as XB acceptors and N-haloimides as XB
donors: N-iodosuccinimide (NIS), N-bromosuccinimide (NBS), N-
iodophthalimide (NIP), N-bromophthalimide (NBP), N-iodosaccharin
(NISac), and N-bromosaccharin (NBSac).

δ
N Xδ + bonds, though being strongly polarized, can
tolerate the high nucleophilicity of N-oxides, and this
enables us to measure the binding affinities of N X··· O N +
motifs also in solution.

Results and Discussion

Computational Studies

In our earlier study on 54 N-halosaccharin-PyNO XB

complexes,[76] we found that combining the local positive
Figure 4. (a) Computed electrostatic potential surface (ESP) at the
maxima of the electrostatic potential (VS,max) of the XB
PBE0-D3/def2-TZVP level projected on the 0.001 au electron density
donor halogen, the atomic charge (Q) and inverse of the surfaces of N-haloimides with VS,max values, NBSac, NBS, NBP (top:
average local ionization energies (1/IS,min) of the XB acceptor left-to-right), and NISac, NIS, and NIP (below: left-to-right). (b) Chart
oxygen of PyNOs allows a simple method of prediction of showing atomic charge (Q) and local ionisation energy (IS,min) of the
XB interaction energies (ΔE). This simple electrostatic pyridine N-oxides’ oxygen atom. Color code: Q (orange bars), IS,min (eV)
model (named here as SiElMo) estimates the ΔE as: (green line + red dots), and pyridine N-oxides (maroon numbers).

Angew. Chem. Int. Ed. 2023, 62, e202307372 (3 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

C6F5I (+ 134 kJ mol 1) and C6F5Br (+ 100 kJ mol 1).[83] The TZVP method does not significantly change the correlation
NIS and NBS have slightly higher Vs,max values than their between the ΔESiElMo and ΔEDFT [Figure S12 (SI Page 14),
corresponding NIP and NBP counterparts with differences R2 = 0.964], demonstrating the robustness of the SiElMo
of 0.8 and 1.9 kJ mol 1, respectively. model.
In the literature, some success has been achieved in Furthermore, similar observations of a strong linear link
describing the influence of the XB acceptor to the XB between SiElMo energies and ΔEDFT are obtained when
strength using minimum value of the surface electrostatic comparing ΔEDFT calculated with the more demanding
potential, Vs,min.[84–86] However, on the NISac and NBSac ωB97X-D/aug-cc-pVTZ(PP) DFT method and SiElMo using
complexes our observation[76] was that the Vs,min values do the original equation R2 = 0.983 [Figure S13 (SI page 14)] or
not lead to best possible correlation with the XB ΔEDFT an equation refitted to the full dataset R2 = 0.984 [Figure S14
values. Instead, the Q value of the N-oxide oxygen (Fig- (SI Page 15)]. Both ΔESiElMo and ΔEDFT rank the interaction
ure 4b) appeared to be a better description of the acceptor’s strengths of N-haloimide complexes as expected: NISac >
role in the formation of halogen bonds. The poor perform- NIS ^ NIP > NBSac > NBS ^ NBP. The simple SiElMo mod-
ance of Vs,min alone can be attributed to the position of the el from the small data set (32 complexes[76]) is able to predict
Vs,min of the oxygen atom, which is not coinciding with the XB interaction energies of all 90 XB complexes remarkably
direction of the halogen bond.[76] A smaller but significant well.
contribution to the XB strengths comes from the reciprocal The SiElMo model’s success in estimating XB interac-
value of the minima of the average local ionization energies, tion energies indicates that ΔE’s are largely influenced by
IS,min (Figure 4b), of the XB acceptor oxygen, that describes the Vs,max of the halogen atom, while being affected to lesser
how easily the XB acceptor molecules are polarized during extent by the Q-value of the N-oxide oxygen. The apparent
the XB formation. Interaction energies for 90 XB complexes strong influence of Vs,max to XB interaction energies stems
were predicted using SiElMo equation[76] (see above) from the difference changing the identity of XB donor atom
together with VS,max, Q, and IS,min values obtained from from iodine to bromine causes. If the changes to XB
PBE0-D3/def2-TZVP calculations and correlated with the interaction energies caused by replacing the N-iodoimide by
XB interaction energies calculated at PBE0-D3/def2-TZVP a different N-iodoimide are compared to the changes caused
level of theory for optimized complex structures as shown in by varying the substituents in PyNOs the maximum changes
Figure 5 (See Supporting Information S2.1 for details). that can be affected are very similar. As an example,
Applying such a simple electrostatic model (SiElMo) a interaction energies ΔEDFT of NISac-1 and NISac-21[76] are
surprisingly good linear correlation between SiElMo esti- 65 and 85 kJ mol 1, while for NIS-1 and NIS-21 they are
mated and PBE0 calculated energies is observed (R2 = 52 and 67 kJ mol 1, respectively. Thus, changing the XB
0.963). Refitting of the SiElMo equation to the full dataset donor in complexes of 1 and 21 changes ΔEDFT by 13 and
of all 90 XB complexes calculated with PBE0-D3/def2- 18 kJ mol 1, respectively. By contrast, with the same com-

Figure 5. Linear correlation between XB interaction energies of 90 XB complexes calculated with the simple electrostatic model ΔESiElMo [= 0.95
(Qoxygen × VS,max)–1760 × 1/IS,min(oxygen) + 220] and those calculated at the PBE0-D3/def2-TZVP level of theory for optimized structures (R2 = 0.963).
Color code: NIS (purple dots), NIP (olive dots), NISac (blue dots), NBS (bright green squares), NBP (pink squares), and NBSac (orange squares)
complexes. The computed ΔEDFT values for NISac/NBSac complexes are taken from our earlier work.[76]

Angew. Chem. Int. Ed. 2023, 62, e202307372 (4 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

plexes the change of XB acceptor in NIS and NISac

complexes changes ΔEDFT by 15 and 20 kJ mol 1, respec-
tively.

X-Ray Crystallography

Seventy-five crystal structures were assessed in order to

understand the halogen bonding properties through the
oxygen atom of the N-oxide, namely the elongation of N X
bond with simultaneous shortening of the N X··· O N +
halogen bond. Close inspection of the X-ray structures
reveals that the asymmetric units’ of 61/75 structures contain
one 1 : 1 XB complex, whereas the others feature deviations
from the 1 : 1 stoichiometry. In these 14 other structures the
N X··· O N + distances are nearly the same as in the 1 : 1
structures, differing only, at maximum, by 0.06 Å, For
example, the asymmetric unit of NIS-3 contains four
crystallographically independent 1 : 1 complexes, whose
N I··· O N + distances are 2.407(3), 2.434(3), 2.443(3), and
2.453(3) Å. As common feature to all the complexes
N X··· O N + contact distance is markedly shorter than the
sum of the van der Waals radii of X- and O-atoms and the Figure 6. (a-d) Halogen and hydrogen monodentate and multifurcated
modes of PyNOs (X = Br/I of imide, H = hydrogen of H C). (e)
contact is associated with elongated imide N X bond, and
Correlation plot of the XB interaction energies calculated at the PBE0-
N X··· O N + angle that is close to 180 degrees [See D3/def2-TZVP level of theory in the crystal-structure geometry DEcrystal
DFT
Tables S16–S21]. For the studied iodoimide-PyNO com- and predicted with the simple model DESiElMo (R2 = 0.906) for 66
plexes the N I··· O N + halogen bond lengths range from monodentate N-haloimide PyNO XB complexes. Color legend: NBS
2.276(2) to 2.584(5) Å and for bromoimide-PyNO complexes (green squares), NBP (pink squares), NBSac (orange squares), NIS
the corresponding N Br··· O N + distances are from (purple dots), NIP (olive dots), and NISac (blue dots).
2.217(2) to 2.509(3) Å. The NIS-20 [2.354(3) Å, RXB = 0.67]
of the NIS series, NIP-2 [2.392(2) Å, RXB = 0.68] for the NIP
and NISac-5 [2.276(2) Å, RXB = 0.65][76] for the NISac been emphasized in crystal engineering studies.[87–91] The
manifest the shortest N I··· O N + XB distances. The NBS-8 asymmetric units in two of the 75 structures, NIS-20 and
[2.34 Å, RXB = 0.69] for NBS, NBP-20 [2.323(4) Å, RXB = NBSac-23, have a second imide that is not involved in
0.69] for NBP and NBSac-20 (2.217(2) Å, RXB = 0.66][76] for N X··· O N + halogen bonding. The second imide N X
NBSac exhibit the shortest Br···O distances. The RXB values forms N X···O=C type XBs to the C=O of the NIS-20 and
of the N-halosuccinimide and N-halosaccharin complexes NBSac-23 complexes [Figure S16 (SI Page 31)].
are only slightly larger than the RHB values of the Given the good performance of the SiElMo model in
corresponding N-imide-PyNO hydrogen-bonded complexes reproducing the ΔEDFT of the optimized structures we chose
[Figure S15 (SI Page 31)]. to test the energies predicted by the SiElMo against
In N H···NPy and N X···NPy complexes the pyridine interaction energies calculated for crystal structure geo-
nitrogen interacts with one donor through only one hydro- metries DEcrystal
DFT with PBE0-D3/def2-TZVP method. The
gen or halogen bond. In contrast, the oxygen in the N-oxide linear correlation between ΔESiElMo and DEcrystal 2
DFT (R = 0.906)
has a multidentate acceptor capability and can interact to demonstrates that the model can be reliably used to predict
either one XB or HB donor (monodentate), one XB and general energy trends in the solid-state structures of N-
HB donor or two XB donors (bidentate) or one XB and two haloimide PyNO XB complexes (Figure 6e). The same level
HB donors (tridentate) simultaneously as depicted in of linear correlation is observed for the DEopt DFT of optimized
Figures 6a–d. The crystal packing analysis of the 75 single structures and DEcrystal
DFT [Figure S9 (SI Page 10)] inferring that
crystal X-ray structures reveal that the N-oxide oxygen the XB are likely to retain their dominant structure directing
manifests the monodentate interaction (one N X··· O N + role in also the solid-state structures.
XB) in only one of the 75 structures. The others exhibit The lengthening of the N X bond occurs upon XB
multidentate interaction patterns with μ2-O,O in 6/75 (two formation [Tables S16–S21 (SI Pages 28–30) and Figure S17
N X··· O N + XBs), μ2-O,O in 48/75 (one N X··· O N + (SI Page 32)]. The imide N X bond elongation defined as,
XB, one C H··· O N + HB), and μ3-O,O,O in 22/75 (one Δ(N X) = (N X)complex–(N X)ligand, is in the range of 0.016–
N X··· O N + XB and two C H··· O N + HBs). Packing 0.05 Å for NBS, 0.018–0.058 Å for NBP, 0.048–0.12 Å for
forces are responsible for the multidentate C H··· O N + NBSac, 0.005–0.056 Å for NIS, 0.05–0.07 Å for NIP and
interactions and, they play a critical role in the stabilization 0.069–0.114 Å for NISac complexes. The Δ(N X) values are
of XB complexes in the solid-state. The significance of clearly larger for donors with larger VS,max (bigger σ-hole). It
C H··· O N + interactions involving PyNOs has previously is generally accepted that in D X···A halogen-bonded

Angew. Chem. Int. Ed. 2023, 62, e202307372 (5 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

complexes, the (D X)complex bond lengths are longer than the oxygen atom of the N-oxide. In their HB study Steiner et al.
(D X)ligand, and the (D X)complex lengthening increases with plotted O H···N and O H···O distances against O···N and
decreasing X···A halogen bond lengths. The complexes with O···O distances in O H···N and O H···O hydrogen-bonded
additional XB and/or HB contacts shown in Figure 6a–d systems. Plotting O···N and O···O distances as a function of
suggest that the additional N X··· O N + and C H··· O N + O H···N and O H···O distances results in a curved path,
interactions could have a major role in stabilizing the crystal where the midpoint of the curve corresponds to symmetric
lattice of the 1 : 1 donor:acceptor XB complexes. If this is the O···H···N and O···H···O bonding.[8] For instance, in the
case, there should not be a direct relationship between N X correlation of O H and H···O distances against O···N
and X···O distances. To investigate this, we compiled and separations, the O H bond continuously elongates with
plotted the N X distances as a function of X···O values decreasing H···O distance until a symmetric geometry for
observed in crystal structures of the XB complexes (Fig- O H N is reached at O···N separation of about ~ 2.50 Å. In
ure 7). this situation, the distance of the H-atom from the N and O-
Overall, the N X elongations and X···O distances with a atoms is 1.25 Å. Analogous methodology can be used to
slope of 3.49 for Br···O and 2.36 for I···O halogen bonds evaluate the symmetry of a N X O halogen-bonded system.
manifest a moderate correlation R2 = 0.866 for the Br and Plots of N X and X···O versus (imide)N···O(PyNO) dis-
R2 = 0.842 for the I complexes. The negative slope indicates tances for N-iodoimide and N-bromoimide complexes,
an inverse relationship, namely the N X bond length respectively, are shown in Figure 8a and 8b. The plots of
increases while the X···O distance decreases. The Br···O and N···O distances as the function of N X and X···O distances
I···O distances group based on the Vs,max values. Stronger forms a parabolic curve where the two values approach a
XBs observed for the NBSac and NISac complexes, show a common minimum with N X=O N. The parabolic curve
distance variation from 2.22 to 2.35 Å and 2.28 to 2.35 Å, reaches minimum at a N···O distance of ~ 4.1 Å for
respectively, forming well separated sets. The Vs,max values bromoimide-PyNO complexes and ~ 4.4 Å for iodoimide-
of bromine in NBS and NBP and of iodine in NIS and NIP PyNO complexes. Although haloimide-N-oxide systems
complexes differ only by ~ 2 kJ mol 1. As a result, the X···O cannot reach this symmetric N X O bonding situation the
distances of the NBS and NBP, and NIS and NIP complexes distances for hypothetical symmetrical system are compara-
are populated in narrow range of 2.37–2.51 Å and 2.38– ble to the distances found in carbonyl hypoiodite [4-
2.46 Å, respectively, and do not form clearly distinguishable bromobenzoate-DMAP-iodine(I)] complex where both
sets as the NBSac and NISac complexes do. Further, the N X and X O bond lengths are 2.22 Å with an N···O
weak and bifurcated I···O contacts manifest themselves as a separation of 4.43 Å (Figure 8c).[50] This is possible as the
separate set for the NISac complexes. The monodentate and pyridinic nitrogen atom can accept the iodine cation,
bifurcated Br···O distances are comparable, and their groups resulting in the unstable iodopyridinium cation (which is
diffuse into one another. The X···O distances seem un- then stabilized by the 4-bromobenzoate anion). The iodine
affected by the non-XB interactions such as C H··· O N + cation can even jump to the pyridinic N atom, as happens in
HBs and depend more on the Vs,max values indicating that trifluoroacetate-DMAP-iodine(I) hypoiodite complex with
the N X··· O N + motifs involving N-haloimides and PyNOs I N 2.173(4) and O···I 2.312 Å and O···N of 4.49 Å (Fig-
are robust in the solid-state. ure 8e).
Unlike in hypoiodite complexes[50] and Barluenga-type
iodine(I) complexes[41,92] it is not possible to achieve a
symmetric N X O halogen-bonded system using N-haloi- Solution Studies
mides and PyNOs, as the iodine(I) cannot “jump” to the
Solution 15N NMR Studies

The computational study of NIS-1, for example, indicates

that the atomic charges on nitrogen of NIS, and nitrogen
and oxygen atoms of 1, become more negative upon XB
complexation (Figure 9), suggesting that monitoring the
NMR changes of oxygen and nitrogen atoms in solution
could be used to evaluate complexation strength. However,
the NMR active oxygen isotope 17O has a natural abundance
of only 0.035 %, which makes the 17O NMR analysis
challenging. On the other hand, the active isotope of
nitrogen, 15N, has a natural abundance of 0.37 %, making it
suitable for 15N NMR study.
The 15N NMR coordination shift, δ15Ncoord, defined as the
difference between the chemical shift of a complexed ligand
(δ15Ncompl) and its uncomplexed form (δ15Nligand), is a useful
method for estimating the strength of coordination bonds. A
Figure 7. Correlation of X···O versus N X distances in the single- larger absolute coordination shift, j Δδ15Ncoord j=δ15Ncompl–
crystalline XB complexes. δ15Nligand indicates a stronger bond. It has previously been

Angew. Chem. Int. Ed. 2023, 62, e202307372 (6 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

Figure 8. Solid-state X-ray crystal structures data. Correlation of (a) N I and I···O versus (imide)N···O(PyNO), and (b) N Br and Br···O versus
(imide)N···O(PyNO). (c) Halogen-bonded symmetric situation of COO ···I N + ion-pair (salt) in an asymmetric (DMAP)N I O(4-bromobenzoate)
moiety, and (d and e) a comparable asymmetric situation of NISac-5 (the strongest XB complex) and COO ···I N + ion-pair (salt) in
(DMAP)N I O (trifluoroacetate) moiety.

Figure 9. Chemdraw representation showing changes in the QTAIM

atomic charges on formation of 1 : 1 XB adduct NIS-1.

used to describe the halogen bonding strengths and confirm

complexations of Barluenga-type [N X N] + [36] and carbonyl
hypoiodite complexes C(O) O I NPy.[50–52] The 15N NMR
data was acquired to determine whether the Δδ15Ncoord of the
nitrogen of PyNO and XB donor would follow the Vs,max
values of the halogens. For NIS and NBS complexes,
determining Δδ15Ncoord was successful, but lack of protons at
the 3-position in other XB donors prevented the determi-
nation in others. The successful estimation of Δδ15Ncoord of
N-oxide nitrogen atom for all PyNOs was achieved and their
Δδ15Ncoord values were used to compare the relative strengths
Figure 10. Summary of solution NMR KXB and Δδ15Ncoord values.
of X···O halogen bonds. The stronger the X···O interaction,
the larger the Δδ15Ncoord of the PyNO nitrogen atom. The
Δδ15Ncoord values of NIS, NIP, NISac, NBS, NBP, and NBSac
complexes are in the ranges of 5.1–15.7 ppm, 3.9–12.6 ppm, uncertainties in ranking the Br···O XBs using the Δδ15Ncoord
8.2–20.2 ppm, 0.3–2.8 ppm, 0.1–4.6 ppm and 2.3–22.2 ppm, values of PyNO nitrogen due to weak binding observed for
respectively (Figure 10). The magnitude of PyNOs’ nitrogen the NBS and NBP complexes (Tables S51–53, Supporting
Δδ15Ncoord values of the N-haloimide-PyNO complexes follow Information pages 207–209).The above findings are also
the sequence of haloimide halogen Vs,max values, NISac > consistent with the observation that as the halogen size
NIS > NIP for the I···O XBs. However, there are significant

Angew. Chem. Int. Ed. 2023, 62, e202307372 (7 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

decreases, it’s XB donor properties (diminishing VS,max 298 K, whereas NBP and NBS complexes could not be
values) also decrease i.e., I > Br > Cl > F.[93] measured due to weak binding. The KXB values of NBS-7
and NBP-7 were measured at 273 K and 253 K to check if
they follow the computed bromine Vs,max values. The NBS-7
Solution 1H NMR Association Constants has a KXB value of 5 and 7 M 1 at 273 K and 253 K,
respectively. Estimation of the KXB for NBP-7 at 273 K was
The scope of our preliminary work[76] was further extended unsuccessful as no chemical shift changes to the donor
to include the solution NMR association constants (KXB) of protons were observed, however binding was observed at
NIS and NIP complexes. By combining the previous[76] and 253 K, being 3 M 1. The slightly larger KXB of NBS-7 than
the current results, plotting the binding constants between NBP-7 clearly suggest that NBS is a stronger donor and
different N-haloimides with the 27 N-oxides allows evalua- NBP is a weaker XB donor, as determined by the computed
tion of the solution behaviour in the N-haloimide-PyNO bromine Vs,max values (Figure 4a).
complexes (Figure 11). The KXB’s were determined from The logarithmic KXB values of NISac, NIS, NIP and
changes in NISac and NBSac proton resonances caused by NBSac complexes, obtained in CDCl3, were plotted for each
the XB complexation. With the use of HypNMR2006[94,95] XB acceptor separately (Figure 11), revealing interesting
and the online Bindfit[96] programs, the KXB values were trends. Figure 11 provides some useful insights: (i) The
determined for a 1 : 1 donor-acceptor model (Table S54, average KXB values of all complexes follow the Vs,max of the
Supporting Information page 210). XB donor. It should be noted that the very high KXB, and
At 298 K the KXB values of NIS complexes in CDCl3 very likely erroneous, value for NISac-7 (> 108) has been
range from 222 to 6434 M 1, and of NIP complexes from 20 excluded from the average. The bromine-atom donor series
to 448 M 1. Titrations of NIS complexes were also carried of NBSac complexes with Br···O XBs shows weak, but
out in XB competing solvent, [D6]acetone, since some of the consistent, binding. (ii) The strongest I···O XBs occurring for
KXB values of NIS complexes in CDCl3 manifest high fitting NISac complexes have a conspicuously random distribution
errors. When compared to CDCl3, the KXB values of NIS of the KXB values, whereas weaker I···O XBs for NIS and
complexes in [D6]acetone are smaller and range from 2 to NIP complexes have a more ‘’steady’’ distribution. This
945 M 1, with small fitting errors. The KXB values of the NIS observation suggests that the simple interpretation of
and NIP complexes were smaller than those for NISac relating the chemical shift changes directly to KXB associa-
complexes, which varied between 1180 and > 108 M 1. Only tion constants might break for the complexes of XB donors
the KXB values of NBSac complexes were measurable at with the largest VS,max values and the observed changes are

Figure 11. Chart displaying KXB values trend of NISac-Z (blue diamonds), NIS-Z (red squares), NIP-Z (green dots), and NBSac-Z (olive triangles).
The KXB value of the NISac-7 is unknown; the blue star indicates the estimated value but is excluded from the average.

Angew. Chem. Int. Ed. 2023, 62, e202307372 (8 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

likely to reflect decomposition and/or influence of secondary and the atomic charge and average local ionization energies
interactions as well. These factors could explain the erratic of the PyNOs’ oxygen. There is a strong correlation (R2 =
pattern as well as large fitting errors to KXB values of NISac 0.963) between ΔESiElMo model energies and the halogen
complexes (Table S54, Supporting Information page 210). bond interaction energies ΔEDFT determined from the DFT
optimized structures of the complexes. Both the ΔESiElMo and
ΔEDFT rank the XB interaction strengths in the order: N-
Categories of Weak, Moderate, Strong N X··· O N + Halogen iodosaccharin > N-iodosuccinimide ^ N-iodophthalimide >
Bonds N-bromosaccharin > N-bromosuccinimide ^ N-bromophtha-
limide. The X···O interaction energies calculated using
We combine our data for helpful practical purposes to crystal structure coordinates (DEcryst
DFT ) and ΔESiElMo show a
introduce indicators for the categories of weak, moderate, strong linear correlation (R2 = 0.902), demonstrating the
and strong XBs, as shown in Table 1. Interaction energies robustness of the SiElMo model and proving that the XB
and X···O distances have no clear boundaries and numerical interaction is the dominant non-covalent interaction motif
data should only be used as a general guide. Additionally, that determines the N-haloimide-PyNO complexes. The
different donor-acceptor partners generate specific non- success of using the SiElMo model, that was developed on a
covalent bonds in solutions and crystals, and they rely on small training set, to predict the XB energies of a much
the functional groups in the donors and acceptors. A notable larger set of complexes paves the way for using the model as
difference between categorization of HBs and XBs is that an expedient way of characterizing X···O halogen bonds in
ffN X···O angles cannot be related to X···O distances in the larger systems such as biomolecules. The finding also raises
same way as weak (> 90°), moderate (> 130°), and strong the question if the model or similar model could be
(170–180°) HBs separated into classes based on the extended for other types of XBs as well.
angles.[97] For example, the strongest I···O of NISac-5 [2.276 In the solid-state, the X···O distances of N-iodoimide-
(2) Å, ffN I···O = 176.14(10)°] shows larger deviation from PyNO complexes vary from 2.276(2) to 2.584(5) Å, whereas
the linearity than the weakest I···O XB of NIS-8 [2.584(5) Å, for N-bromoimide-PyNO complexes they range from
ffN I···O = 178.6(3)°], even though linearity is what is 2.217(2) to 2.509(3) Å. The correlation between (imide)N X
expected of strong XBs. The X-atom preference for high vs X···O distances establishes a common link, that is, as the
directionality can be attributed for the near linearity in all N X bond lengthens, the X···O distance shortens. Within
complexes [167.72(15)°–179.38(14)°, Tables S16–S21, Sup- this correlation, the X···O halogen bonds form groups that
porting Information pages 28–30], while packing forces are can be ranked based on donor halogen sigma-hole strength,
expected to account for the small variations in the bond N-halosaccharin > N-halosuccinimide ^ N-halophthalimide.
angles. N-iodosaccharin complexes have short X···O distances, with
normalized interaction ratios (RXB) ranging from 0.65 to
0.70, and hence are the shortest N I··· O N + halogen bonds
Conclusion of their class. These values are only minutely larger than the
RXB values of 0.64 observed for [bis(pyridine)iodine] + XB
Six N-haloimide XB donors and 27 pyridine N-oxide complexes. Out of 75 crystal structures, in only one
(PyNO) XB acceptors are studied for N X··· O N + (X = I, structure, the N-oxide oxygen has a monodentate
Br) halogen bonding interactions In silico, crystals, and N X··· O N + XB bonding mode and the others exhibit
solution with the aim of understanding their bonding nature. multidentate interaction patterns with μ2-O,O in 6/75 (two
DFT XB interaction energies for N-iodoimide-PyNO com- N X··· O N + XBs), μ2-O,O in 48/75 (one N X··· O N + XB,
plexes range from 56 to 120 kJ mol 1, while for N- one C H··· O N + HB), and μ3-O,O,O in 22/75 (one
bromoimide-PyNO complexes, they are between 38 and N X··· O N + XB and two C H··· O N + HBs). This finding
82 kJ mol 1. These bond energies are the strongest of their shows that the N-oxide oxygen‘s polydentate coordination
kind but ^ 40 kJmol 1 weaker than the well-known + X···N nature plays a significant role in crystal packing processes
XBs in [N X N] + halogen-bonded systems. A simple and that in solution, the N-oxide oxygen may have a
electrostatic model (SiElMo) is presented to predict the comparable or even higher tendency for polydenticity. In
X···O XB bond energies from the properties of XB donor solution, the binding affinities can be significantly influenced
halogen and acceptor oxygen atoms. SiElMo uses the by the solvation and entropy of the donor and acceptor
positive electrostatic surface potential of the donor halogen molecules, which neither DFT nor crystal structures reflect.
Although desirable, correlations between DFT/solid-state
and solution are not seen. Solution NMR data are treated
Table 1: Classification of strong, moderate, and weak N X··· O N + separately. Trends for NMR association constants have been
XBs. calculated independently for each of the N-haloimide-PyNO
families. N-iodosaccharin complexes with strong XB donor
weak moderate Strong
capacities exhibit erratic patterns of association constant
X···O Bond length (Å) > 2.5 2.4–2.5 < 2.3 distributions, whereas N-haloimide complexes with weak
N X Lengthening < 0.05 0.05–0.09 > 0.10 donor properties have more ‘’steady’’ distributions. The
Bond energy (kJ mol 1) < 20 20–60 > 60 association constants follow the sigma-hole strengths of XB
Association constants (M 1) 1–99 100–1000 > 1000

Angew. Chem. Int. Ed. 2023, 62, e202307372 (9 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

donors, which is in agreement with DFT and solid-state X- [13] G. R. Desiraju, P. Shing Ho, L. Kloo, A. C. Legon, R.
ray crystallography data. Marquardt, P. Metrangolo, P. Politzer, G. Resnati, K. Rissa-
nen, Pure Appl. Chem. 2013, 85, 1711–1713.
[14] E. Arunan, G. R. Desiraju, R. A. Klein, J. Sadlej, S. Scheiner,
I. Alkorta, D. C. Clary, R. H. Crabtree, J. J. Dannenber, P.
Social media promotion Hobza, H. G. Kjaergaard, A. C. Legon, B. Mennucci, D. J.
Nesbitt, Pure Appl. Chem. 2011, 83, 1637–1641.
Halogen-bond energies were evaluated and classified using [15] K. Rissanen, CrystEngComm 2008, 10, 1107–1113.
N-haloimides as halogen-bond donors and aromatic N- [16] K. E. Riley, J. S. Murray, J. Fanfrlík, J. Řezáč, R. J. Solá, M. C.
oxides as halogen-bond acceptors. The sigma-hole of the Concha, F. M. Ramos, P. Politzer, J. Mol. Model. 2011, 17,
halogen in N-haloimides allows the quantification the bond 3309–3318.
strengths, and it is surprising how little oxygen impacts [17] T. Clark, M. Hennemann, J. S. Murray, P. Politzer, J. Mol.
strengths. Model. 2007, 13, 291–296.
[18] P. Politzer, J. S. Murray, T. Clark, G. Resnati, Phys. Chem.
Chem. Phys. 2017, 19, 32166–32178.
[19] Q. M. Dang, J. H. Simpson, C. A. Parish, M. C. Leopold, J.
Acknowledgements Phys. Chem. A 2021, 125, 9377–9393.
[20] P. Metrangolo, G. Resnati, Halogen Bonding: Fundamentals
The authors gratefully acknowledge financial support from and Applications, Springer-Verlag, Berlin, 2008.
the Academy of Finland (RP grant no. 298817, KR grant no. [21] P. Metrangolo, G. Resnati, Halogen Bonding: Impact on
351121), and provisions of computational resources by the Material Chemistry and Life Sciences, Springer International
Finnish Grid and Cloud Infrastructure (urn:nbn:fi:research- Publishing, Cham, 2015.
[22] C. B. Aakeroy, D. L. Bryce, G. R. Desiraju, A. Frontera, A. C.
infras-2016072533) and Prof. Dr Heikki Tuononen (Univer-
Legon, F. Nicotra, K. Rissanen, S. Scheiner, G. Terraneo, P.
sity of Jyväskylä). Metrangolo, G. Resnati, Pure Appl. Chem. 2019, 91, 1889–
1892.
[23] A. Bauzá, A. Frontera, Angew. Chem. Int. Ed. 2015, 54, 7340–
Conflict of Interest 7343.
[24] A. Bauzá, I. Alkorta, J. Elguero, T. J. Mooibroek, A. Frontera,
The authors declare no conflict of interest. Angew. Chem. Int. Ed. 2020, 59, 17482–17487.
[25] L. Brammer, Faraday Discuss. 2017, 203, 485–507.
[26] S. J. Grabowski, Coord. Chem. Rev. 2020, 407, 213171.
[27] L. Vogel, P. Wonner, S. M. Huber, Angew. Chem. Int. Ed.
Data Availability Statement 2019, 58, 1880–1891.
[28] M. S. Taylor, Coord. Chem. Rev. 2020, 413, 213270.
The data that support the findings of this study are available [29] K. T. Mahmudov, A. V. Gurbanov, M. F. C. Guedes da Silva,
from the corresponding author upon reasonable request. A. J. L. Pombeiro, in Noncovalent Interactions in Catalysis, The
Royal Society of Chemistry, Croydon, 2019.
[30] Noncovalent Forces (Ed.: S. Scheiner), Springer International
Keywords: DFT · Halogen Bond · N-Haloimide · N-Oxide ·
Publishing, Cham, 2015.
X-Ray Structure [31] J. Zheng, A. Suwardi, C. J. E. Wong, X. J. Loh, Z. Li, Nano-
scale Adv. 2021, 3, 6342–6357.
[32] G. Espuña, G. Arsequell, G. Valencia, J. Barluenga, J. M.
Alvarez-Gutiérrez, A. Ballesteros, J. M. González, Angew.
Chem. Int. Ed. 2004, 43, 325–329.
[33] J. Barluenga, H. Vázquez-Villa, A. Ballesteros, J. M. González,
[1] M. Raynal, P. Ballester, A. Vidal-Ferran, P. W. N. M.
J. Am. Chem. Soc. 2003, 125, 9028–9029.
Van Leeuwen, Chem. Soc. Rev. 2014, 43, 1660–1733.
[34] J. Barluenga, M. Trincado, E. Rubio, J. M. González, Angew.
[2] O. Yamauchi, Phys. Sci. Rev. 2019, 1, 20160001.
Chem. Int. Ed. 2003, 42, 2406–2409.
[3] S. Jena, J. Dutta, K. D. Tulsiyan, A. K. Sahu, S. S. Choudhury,
[35] J. Barluenga, F. González-Bobes, M. C. Murguía, S. R. Anan-
H. S. Biswal, Chem. Soc. Rev. 2022, 51, 4261–4286.
thoju, J. M. González, Chem. Eur. J. 2004, 10, 4206–4213.
[4] B. Alberts, Molecular Biology of the Cell, Garland Science,
New York, 2002. [36] L. Turunen, M. Erdélyi, Chem. Soc. Rev. 2020, 49, 2688–2700.
[5] K. Müller-Dethlefs, P. Hobza, Chem. Rev. 2000, 100, 143–167. [37] L. Turunen, A. Peuronen, S. Forsblom, E. Kalenius, M.
[6] G. R. Desiraju, T. Steiner, The Weak Hydrogen Bond: In Lahtinen, K. Rissanen, Chem. Eur. J. 2017, 23, 11714–11718.
Structural Chemistry and Biology, Oxford University Press, [38] L. Turunen, U. Warzok, C. A. Schalley, K. Rissanen, Chem
2001. 2017, 3, 861–869.
[7] M. C. Etter, Acc. Chem. Res. 1990, 23, 120–126. [39] A. Vanderkooy, A. K. Gupta, T. Földes, S. Lindblad, A.
[8] T. Steiner, Angew. Chem. Int. Ed. 2002, 41, 48–76. Orthaber, I. Pápai, M. Erdélyi, Angew. Chem. Int. Ed. 2019, 58,
[9] S. J. Grabowski, Chem. Rev. 2011, 111, 2597–2625. 9012–9016.
[10] Hydrogen Bonding in Organic Synthesis (Ed.: Petri M. Pihko), [40] G. Gong, S. Lv, J. Han, F. Xie, Q. Li, N. Xia, W. Zeng, Y.
Wiley-VCH, Weinheim, 2009. Chen, L. Wang, J. Wang, S. Chen, Angew. Chem. Int. Ed. 2021,
[11] S.-W. Kuo, Hydrogen Bonding in Polymeric Materials, Wiley- 60, 14831–14835.
VCH, Weinheim, 2018. [41] A.-C. C. Carlsson, J. Gräfenstein, J. L. Laurila, J. Bergquist, M.
[12] S. J. D. Lugger, S. J. A. Houben, Y. Foelen, M. G. Debije, Erdélyi, Chem. Commun. 2012, 48, 1458–1460.
A. P. H. J. Schenning, D. J. Mulder, Chem. Rev. 2022, 122, [42] M. Bedin, A. Karim, M. Reitti, A.-C. C. Carlsson, F. Topić, M.
4946–4975. Cetina, F. Pan, V. Havel, F. Al-Ameri, V. Sindelar, K.

Angew. Chem. Int. Ed. 2023, 62, e202307372 (10 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
 15213773, 2023, 34, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202307372, Wiley Online Library on [01/07/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Angewandte
Research Articles Chemie

Rissanen, J. Grafenstein, M. Erdelyi, Chem. Sci. 2015, 6, 3746– [72] R. Puttreddy, F. Topić, A. Valkonen, K. Rissanen, Crystals
3756. (Basel) 2017, 7, 214.
[43] J. S. Ward, G. Fiorini, A. Frontera, K. Rissanen, Chem. [73] G. R. Hanson, P. Jensen, J. McMurtrie, L. Rintoul, A. S.
Commun. 2020, 56, 8428–8431. Micallef, Chem. Eur. J. 2009, 15, 4156–4164.
[44] D. von der Heiden, K. Rissanen, M. Erdélyi, Chem. Commun. [74] R. Puttreddy, O. Jurcek, S. Bhowmik, T. Mäkelä, K. Rissanen,
2020, 56, 14431–14434. Chem. Commun. 2016, 52, 2338–2341.
[45] S. Yu, J. S. Ward, Dalton Trans. 2022, 51, 4668–4674. [75] R. Puttreddy, N. K. Beyeh, K. Rissanen, CrystEngComm 2016,
[46] J. S. Ward, CrystEngComm 2022, 24, 7029–7033. 18, 793–799.
[47] S. Yu, P. Kumar, J. S. Ward, A. Frontera, K. Rissanen, Chem [76] R. Puttreddy, J. M. Rautiainen, T. Mäkelä, K. Rissanen,
2021, 7, 948–958. Angew. Chem. Int. Ed. 2019, 58, 18610–18618.
[48] J. S. Ward, A. Frontera, K. Rissanen, Inorg. Chem. 2021, 60, [77] O. Makhotkina, J. Lieffrig, O. Jeannin, M. Fourmigué, E.
5383–5390. Aubert, E. Espinosa, Cryst. Growth Des. 2015, 15, 3464–3473.
[49] S. Wilcox, D. Sethio, J. S. Ward, A. Frontera, R. Lindh, K. [78] E. Aubert, E. Espinosa, I. Nicolas, O. Jeannin, M. Fourmigué,
Rissanen, M. Erdélyi, Chem. Commun. 2022, 58, 4977–4980. Faraday Discuss. 2017, 203, 389–406.
[50] S. Yu, J. S. Ward, K. N. Truong, K. Rissanen, Angew. Chem. [79] V. Stilinović, G. Horvat, T. Hrenar, V. Nemec, D. Cinčić,
Int. Ed. 2021, 60, 20739–20743. Chem. Eur. J. 2017, 23, 5244–5257.
[51] E. Kramer, S. Yu, J. S. Ward, K. Rissanen, Dalton Trans. 2021, [80] J. Mavračić, D. Cinčić, B. Kaitner, CrystEngComm 2016, 18,
50, 14990–14993. 3343–3346.
[52] J. S. Ward, J. Martõnova, L. M. E. Wilson, E. Kramer, R. Aav, [81] M. Eraković, D. Cinčić, K. Molčanov, V. Stilinović, Angew.
K. Rissanen, Dalton Trans. 2022, 51, 14646–14653. Chem. Int. Ed. 2019, 131, 15702–15706.
[53] M. Mattila, K. Rissanen, J. S. Ward, Chem. Commun. 2023, 59, [82] I. Castellote, M. Morón, C. Burgos, J. Alvarez-Builla, A.
4648–4651. Martin, P. Gómez-Sal, J. J. Vaquero, Chem. Commun. 2007,
[54] R. Wilcken, M. O. Zimmermann, A. Lange, A. C. Joerger, 1281–1283.
F. M. Boeckler, J. Med. Chem. 2013, 56, 1363–1388. [83] K. N. Truong, J. M. Rautiainen, K. Rissanen, R. Puttreddy,
[55] M. R. Scholfield, C. M. Vander Zanden, M. Carter, P. S. Ho, Cryst. Growth Des. 2020, 20, 5330–5337.
Protein Sci. 2013, 22, 139–152. [84] I. Alkorta, G. Sánchez-Sanz, J. Elguero, CrystEngComm 2013,
[56] M. T. Messina, P. Metrangolo, W. Panzeri, T. Pilati, G. 15, 3178–3186.
Resnati, Tetrahedron 2001, 57, 8543–8550. [85] P. Politzer, J. S. Murray, T. Clark, Phys. Chem. Chem. Phys.
[57] C. B. Aakeröy, T. K. Wijethunga, J. Desper, CrystEngComm 2013, 15, 11178–11189.
2014, 16, 28–31. [86] J. S. Murray, P. Politzer, Crystals 2020, 10, 76.
[58] W. Borley, B. Watson, Y. P. Nizhnik, M. Zeller, S. V. Rosokha, [87] A. V. Shishkina, V. V. Zhurov, A. I. Stash, M. V. Vener, A. A.
J. Phys. Chem. A 2019, 123, 7113–7123. Pinkerton, V. G. Tsirelson, Cryst. Growth Des. 2013, 13, 816–
[59] N. Bedeković, L. Fotović, V. Stilinović, D. Cinčić, Cryst. 828.
Growth Des. 2022, 22, 987–992. [88] A. J. Rybarczyk-Pirek, M. Łukomska-Rogala, K. Wzgarda-Raj,
[60] M. Benito, Y. Roselló, M. Barceló-Oliver, A. Frontera, E. S. Wojtulewski, M. Palusiak, Cryst. Growth Des. 2018, 18,
Molins, Int. J. Mol. Sci. 2021, 22, 10663. 7373–7382.
[61] V. Nemec, D. Cinčić, Cryst. Growth Des. 2022, 22, 5796–5801. [89] L. M. Foroughi, A. J. Matzger, Cryst. Growth Des. 2021, 21,
[62] K. M. Bairagi, K. S. Ingle, R. Bhowal, S. A. Mohurle, A. 5873–5879.
Hasija, O. I. Alwassil, K. N. Venugopala, D. Chopra, S. K. [90] N. J. Babu, L. S. Reddy, A. Nangia, Mol. Pharm. 2007, 4, 417–
Nayak, ChemPlusChem 2021, 86, 1167–1176. 434.
[63] V. V. Panikkattu, A. S. Huber, A. S. Sinha, B. B. Averkiev, [91] L. S. Reddy, N. J. Babu, A. Nangia, Chem. Commun. 2006,
C. B. Aakeröy, Cryst. Growth Des. 2022, 22, 1538–1542. 1369–1371.
[64] V. Nemec, T. Piteša, T. Friščić, D. Cinčić, Cryst. Growth Des. [92] S. B. Hakkert, M. Erdélyi, J. Phys. Org. Chem. 2015, 28, 226–
2020, 20, 3617–3624. 233.
[65] W. X. Wu, H. Wang, W. J. Jin, Cryst. Growth Des. 2018, 18, [93] P. J. Costa, Phys. Sci. Rev. 2017, 2, 20170136.
6742–6747. [94] C. Frassineti, L. Alderighi, P. Gans, A. Sabatini, A. Vacca, S.
[66] W. X. Wu, H. C. Liu, W. J. Jin, Chem. Eur. J. 2022, 28, Ghelli, Anal. Bioanal. Chem. 2003, 376, 1041–1052.
e202103336. [95] C. Frassineti, S. Ghelli, P. Gans, A. Sabatini, M. S. Moruzzi, A.
[67] C. B. Aakeröy, T. K. Wijethunga, J. Desper, CrystEngComm Vacca, Anal. Bioanal. Chem. 1995, 231, 374–382.
2014, 16, 28–31. [96] P. Thordarson, Chem. Soc. Rev. 2011, 40, 1305–1323.
[68] W. X. Wu, H. Wang, W. J. Jin, CrystEngComm 2020, 22, 5649– [97] G. Gilli, P. Gilli, The Nature of the Hydrogen Bond: Outline of
5655. a Comprehensive Hydrogen Bond Theory, Oxford University
[69] C. Cavallotti, P. Metrangolo, F. Meyer, F. Recupero, G. Press, Oxford, 2009.
Resnati, J. Phys. Chem. A 2008, 112, 9911–9918.
[70] Y. P. Nizhnik, A. Sons, M. Zeller, S. V. Rosokha, Cryst.
Growth Des. 2018, 18, 1198–1207. Manuscript received: May 25, 2023
[71] R. Liu, H. Wang, W. J. Jin, Cryst Growth Des 2017, 17, 3331– Accepted manuscript online: June 14, 2023
3337. Version of record online: June 14, 2023

Angew. Chem. Int. Ed. 2023, 62, e202307372 (11 of 11) © 2023 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH