A crystallographic and solid-state NMR study of 1,4-disubstituted 2,5-diketopiperazines

Article abstract of DOI:10.1016/j.molstruc.2021.130157

p-Disubstituted phenyldiketopiperazines 1 (R = H), 2 (R = NO₂) and 3 (R = -N(CH₃)₂) were synthesized and characterized by NMR in solution and IR spectroscopy. The identity of the compounds was confirmed and their fragmenta

Full text of DOI:10.1016/j.molstruc.2021.130157

Journal of Molecular Structure 1234 (2021) 130157
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstr
A crystallographic and solid-state NMR study of 1,4-disubstituted
2
,5-diketopiperazines
a,∗
a
b
c
Berislav Peri c´ , Natalija Pantalon Juraj , Gábor Szalontai , Suzana R. Veli cˇ kovi c´ ,
c
a,d
a,∗
Filip M. Veljkovi c´ , Dražen Viki c´ -Topi c´ , Sre c´ ko I. Kirin
a
Rud¯er Boškovi ´c Institute, Bijeni cˇ ka cesta 54, HR-10000 Zagreb, Croatia
b
NMR Laboratory, University of Pannonia, Egyetem utca 10, H-8201 Veszprém, Hungary
c
ˇ
Department of Physical Chemistry, “VINCA” Institute of Nuclear Sciences - National Institute of thе Republic of Serbia, University of Belgrade, Mike
Petrovi ´c a Alasa 12-14, RS-11000 Belgrade, Serbia
d
Department of Natural and Health Studies, Juraj Dobrila University of Pula, Zagreba cˇ ka 30, HR-52100 Pula, Croatia
a r t i c l e i n f o
a b s t r a c t
Article history:
p-Disubstituted phenyldiketopiperazines 1 (R = H), 2 (R = NO ) and 3 (R = -N(CH ) ) were synthesized
2
3 2
Received 9 November 2020
Revised 8 February 2021
Accepted 10 February 2021
Available online 26 February 2021
and characterized by NMR in solution and IR spectroscopy. The identity of the compounds was confirmed
and their fragmentation analyzed by ESI-MS and HRMS spectrometry. X-ray single crystal structures re-
vealed that the three compounds crystallize in space groups P21 (1), Pbca (2) and P21/c (3), respectively.
The solid-state structures of 1–3 were further analyzed by a combination of solid-state NMR spectroscopy
and GIPAW calculations. The NMR crystallography approach was used to analyze symmetry breaking of
nearly centrosymmetric molecules in 1, and disorder of piperazine groups in crystal structure of 3.
Keywords:
NMR crystallography
2
,5-diketopiperazines
©
2021 Elsevier B.V. All rights reserved.
X-ray diffraction
GIPAW calculations
Solid-state NMR
1
. Introduction
dipolar coupling-chemical shift tensor correlation solid-state NMR.
Wabnitz et al. [11] described the negative ion mass spectra of de-
protonated 2,5-diketopiperazines.
Diketopiperazines are a class of cyclic compounds contain-
ing two amide groups. Depending on the position of the car-
bonyl groups, 2,3-, 2,5- and 2,6- isomers are possible (Fig. 1). 2,5-
Diketopiperazines can be obtained by the condensation of two
α-amino acids and are considered the simplest class of cyclic
peptides [1]. Due to their facile synthesis and their structure
that enables them to bind various receptors, they are exten-
sively studied in pharmaceutical chemistry, [2,3] for example, 2,5-
diketopiperazines prepared starting from indole carboxylic acid
were found to be DNA binding agents with potent cytotoxic ac-
tivity [4]. Tetraketopiperazines that have phenyl substituents on
the amide nitrogen atoms were studied as electrode material for
electrochemical energy storage, [5,6] as electron-donor initiators
of coupling reactions [7] and as antileishmanial chemotherapeutic
agents [8].
In this publication we present the preparation and characteriza-
tion of three 2,5-diketopiperazines 1–3, with different substituents
on the para position of the phenyl rings, (Scheme 1). In addition to
the unsubstituted derivative (1), the nitro (2) and dimethylamino
(3) substituents were chosen as examples of an electron donat-
ing and electron withdrawing group, with σp Hammet constants of
−0.830 and +0.778, respectively. Groups of different electronic in-
fluence, as well as the non-substituted phenyl ring were chosen to
observe the influence of the electronic properties of the molecule,
on the solid-state conformation. The emphasis of this study is the
structural analysis using the combined methodology of single crys-
tal analysis, solid-state NMR measurements and DFT computations
(NMR-crystallography) [12,13]. In particular, the structural analysis
was performed with emphasis on symmetry breaking [14,15] and
disorder effects [16] observed in two structures.
The six membered rings of 2,5-diketopiperazines can exist in
planar or puckered boat forms, and the two forms differ only
slightly in energy [1,9]. Mukhopadhyay et al. [10] studied the
peptide bond cis-trans conformation in 2,5-diketopiperazine using
2. Materials and methods
2
.1. General remarks
∗
Reactions were carried out in ordinary glassware and chemicals
Corresponding authors.
E-mail address: berislav.peric@irb.hr (B. Peri c´ ).
were used as purchased from commercial suppliers without fur-
https://doi.org/10.1016/j.molstruc.2021.130157
0
022-2860/© 2021 Elsevier B.V. All rights reserved.

B. Peri ´c , N.P. Juraj, G. Szalontai et al.
Journal of Molecular Structure 1234 (2021) 130157
of the non-protonated carbons and the methyl groups appear in
the CPNQS spectra. Chemical shift calibration: to external l-alanine
methyl signal = 20.5 ppm relative to TMS. Samples used in ss-
NMR measurements were checked by powder X-ray diffraction on
a PANalytical Aeris instrument; [conditions Bragg-Brentano geom-
˚
etry (ꢀ–2ꢀ), source Cu-Kα (λ = 1.5418 A), measurement from 5°
to 70° (2ꢀ), with 5.2°/min (0.0216° step and 0.25 s/step)] (Figures
S13-S15 in Supplementary material).
Fig. 1. 2,3-, 2,5- and 2,6- isomers of diketopiperazines.
2
.2. Synthetic procedures
ther purification. Reactions were monitored by TLC on Silica Gel
The syntheses of the chloroacetamide precursors P1−P3 are de-
6
0 F254 plates and detected with a UV lamp (254 nm).
scribed in the Supporting Information [17,18].
NMR spectra in solution were obtained on a Bruker Avance
Synthesis of 1,4-diphenylpiperazine-2,5-diones 1–3, general
procedure [18]. A mixture of chloroacetamide (1 eq) and NaOH
(2 eq) in acetonitrile (50 mL) was stirred in a round bottom flask
and refluxed for 6 h. The solvent was evaporated in a vacuum, the
residue suspended in ethyl acetate and washed with water. The
suspension was filtered, the filter cake washed with water and col-
lected, affording 1,4-diphenylpiperazine-2,5-diones (1, 2 or 3).
3
00 or 600 spectrometer, operating at 300 or 600 MHz for 1
H
and on Bruker Avance 600 spectrometer operating at 150 MHz for
13C. If not mentioned otherwise, the spectra are recorded at room
temperature (Figures S1-S6 in Supplementary material). Chemical
shifts, δ (ppm), indicate a downfield shift from the residual sol-
vent signal (1.94 ppm CD CN, 2.50 ppm DMSO–d for 1H NMR,
3
6
1
18.26 ppm CD CN or 39.52 ppm DMSO–d6 for 13C NMR).
1,4-diphenylpiperazine-2,5-dione,
phenylacetamide, P1 (339.2 mg,
4 mmol). Yield: 117.8 mg (0.44 mmol, 44%), white powder,
1.
2-Chloro-N-
3
The mass spectrometric measurements were recorded either in
2 mmol), NaOH (160.0 mg,
the positive or negative ion mode using an Agilent Technologies
6
420 Triple Quad LC/MS (Agilent, USA) equipped for electrospray
Rf = 0.31 (2% methanol in dichloromethane). 1H NMR (600 MHz,
ionization source. The operating conditions for ESI were: nebulizer
gas (nitrogen) flow rate 0.2 mL/min; drying gas (nitrogen) flow rate
DMSO–d ) δ/ppm 7.47–7.41 (m, 8H, o-, m-Ph), 7.33–7.29 (m, 2H,
6
p-Ph), 4.51 (s, 4H, CH ). APT 13C NMR (151 MHz, DMSO–d ) δ/ppm
2
6
8
L/min (300 °C), the spray needle voltage was set to 3.5 kV (“cap-
164.06 (CO), 140.16 (Ph), 128.89 (Ph), 126.67 (Ph), 125.23 (Ph),
+
illary voltage”), the potential between the sampling cone and the
skimmer (“cone voltage”) was set to 135 V. The temperature in the
ion source was kept at 100 °C. Spectra were acquired at a step
size of 0.1 amu over a mass range of 10–2250 amu with a scan
time of 500 ms. Samples were dissolved in water–methanol sol-
vent 50:50 (v/v), at a concentration of 10 mg/mL, and the solutions
were directly infused into the electrospray source. The sample so-
lutions were introduced with a syringe pump at a flow rate of
52.76 (CH ). ESI-MS (m/z): 267.2 ([M+H] , 100%). MALDI-HRMS
2
+
(m/z): calcd for C16
H
N O 267.1128 [M+H] , found 267.1120. IR
14
2
2
(KBr) ν˜ /cm-1: 3473, 3059, 2947, 1654, 1592, 1497, 1469, 1450,
1432, 1317, 1252, 1142, 756, 692, 517, 495. Colorless prism-like
crystals suitable for single-crystal X-ray diffraction were obtained
from deuterated acetonitrile.
1,4-bis(4-nitrophenyl)piperazine-2,5-dione, 2. 2-Chloro-N-(4-
nitrophenyl)acetamide, P2 (429.2 mg, 2 mmol), NaOH (160.0 mg,
4 mmol). Yield: 153.3 mg (0.43 mmol, 43%), yellow powder,
Rf = 0.29 (2% methanol in dichloromethane). 1H NMR (600 MHz,
0.2 mL/min.
The MALDI-TOF/TOF mass spectra were obtained in the posi-
tive ion reflectron mode. Mass spectra of the compounds 1, 2 and
were accumulated using 50, 600 and 100 laser shots, respec-
DMSO–d ) δ/ppm 8.33–8.29 (m, 4H, Ph), 7.77 –7.73 (m, 4H, Ph),
6
3
4.70 (s, 4H, CH ). 13C NMR (151 MHz, DMSO–d ) δ/ppm 164.43
2
6
tively. All samples were analysed using CHCA (Sigma) as a matrix.
The CHCA matrix was prepared in a mixture of water-acetonitrile
(CO), 145.43 (Ph), 144.67 (Ph), 125.00 (Ph), 124.11 (Ph), 52.07 (CH ).
2
+
MALDI-HRMS (m/z): calcd for C16 H12N O 357.0830 [M+H] , found
4
6
5
0:50 (v/v) at a concentration of 10 mg/mL. The samples were pre-
357.0830. IR (KBr) ν˜ /cm-1: 3452, 3105, 3075, 1669, 1590, 1517,
1490, 1348, 1265, 1149, 857, 830, 733, 693, 603, 556, 546, 502,
411. Light yellow prism-like crystals suitable for single-crystal X-
ray diffraction were obtained by vapor diffusion of diethyl ether
into DMSO.
pared in the same solvent. Approximately 1 μL of the mixture
sample/matrix was deposited on the target plate (stainless steel)
and dried at room temperature.
IR spectra were recorded using KBr pellets with a Bruker Alpha
−1
FT-IR spectrometer, in the 4000–350 cm region (Figures S10-S12
in Supplementary material).
1,4-bis(4-(dimethylamino)phenyl)piperazine-2,5-dione, 3. 2-
Chloro-N-(4-(dimethylamino)-phenyl)acetamide, P3 (425.4 mg,
2 mmol), NaOH (160.0 mg, 4 mmol). The organic layer was evap-
orated, washed with acetone and combined with the filter cake.
Solid-state NMR (ssNMR) spectra were recorded on a Bruker
Avance II 400 spectrometer at Larmor frequencies of 400.13, 100.6,
and 40.56 MHz for 1H, 13C, and 15N, respectively; 4 mm MAS
probe using Zirconia rotors, rotation frequency: 5000–12,000 Hz.
Standard Bruker pulse sequences were used: 13C-, and 15N-CPMAS
Yield: 224.89 mg (0.64 mmol, 64%), brown powder, Rf = 0.18 (2%
1
methanol in dichloromethane). H NMR (600 MHz, CD CN) δ/ppm
3
7.19– 7.15 (m, 4H, Ph), 6.80–6.75 (m, 4H, Ph), 4.31 (s, 4H, CH ),
2
vacp, contact times 0.5–2 ms, relaxation delays 3–5 s), 13C CP-
2.94 (s, 12H, CH ). C NMR (151 MHz, CD CN) δ/ppm 165.27 (CO),
13
(
3
3
MAS with non-quaternary suppression (CPNQS), 13C MAS with
150.84 (Ph), 130.64 (Ph), 127.52 (Ph), 113.49 (Ph), 54.55 (CH ),
2
+
+
high-power proton decoupling (HPDEC). Notice that only signals
40.80 (CH ). ESI-MS (m/z): 705.5 [2M+H] , 353.3 ([M+H] , 100%).
3
Scheme 1. Synthesis of 1,4-diphenylpiperazine-2,5-diones 1–3. Reaction conditions: (a) NaOH, CH3CN.
2

B. Peri ´c , N.P. Juraj, G. Szalontai et al.
Journal of Molecular Structure 1234 (2021) 130157
+
MALDI-HRMS (m/z): calcd for C20H24N O 353.1972 [M+H] , found
2.4. DFT calculations
4
2
3
53.1964. IR (KBr) ν˜ /cm-1: 3443, 2890, 2799, 1664, 1608, 1520,
1
463, 1348, 1318, 1125, 1062, 942, 812, 536, 496, 415. Light brown
Structural models of 1, 2 and 3 obtained by single crystal
X-ray diffraction were optimized using the DFT pseudopotential
plane-wave method by Quantum Espresso program [22]. For 3, this
means optimization of two structural models, 3A and 3B, corre-
sponding to two disordered components of piperazine ring orienta-
tion. All atoms were relaxed during optimization and unit cell pa-
rameters were fixed to the values obtained from the X-ray diffrac-
tion. After testing the energy convergence, the cut-off energy of the
plane-waves was set to 90 Ry. The grid of sampling points in the
reciprocal space was 5 × 4 × 2, 2 × 4 × 2 and 4 × 5 × 2, for 1,
prism-like crystals suitable for single-crystal X-ray diffraction were
obtained from acetonitrile.
2
.3. Single crystal X-ray diffraction
Intensity data were collected using an Oxford diffraction Xcal-
˚
ibur CCD diffractometer with Cu-Kα (λ = 1.54184 A) radiation
at room temperature (293 K). Data were processed with Crysal-
isPro program [19] (unit cell determination and data reduction).
The structures were solved by direct methods using SXELXT pro-
gram [20] and refined against F2 on all data by full-matrix least
squares procedure using SHELXL program [21]. All non-hydrogen
atoms were refined with anisotropic displacement parameters. Hy-
drogen atoms were included in the model at the geometrically cal-
culated positions and refined using a riding model. Torsional angles
around C–CH3 bonds of compound 3 were refined (HFIX 137). Also,
for this compound, peaks from difference Fourier electron density
map showed two alternative orientations of the piperazine ring
with respect to the phenyl rings. All atoms from piperazine O=C–
2
and 3 (i.e. 3A and 3B), respectively. Exchange/correlation energy
was calculated using appropriate pseudo-potentials based on the
PBE functional [23]. These pseudopotentials, which contained GI-
PAW reconstruction, were taken from Ceresoli [24]. NMR shielding
tenors and parameters for quadrupolar interaction (electric filed
gradients) were calculated by the gipaw.x module of the Quantum
Espresso program [25]. Optimized coordinates and calculated NMR
parameters for all atoms are deposited in Supplementary material.
3. Results and discussion
CH groups were split across two positions in accordance with the
2
observed difference Fourier map. Restraints that all atoms in in-
dividual disordered parts A and B, share the same occupancy and
that the sum of these occupancies is equal to 1 are used in the
least-square refinements. Almost equal occupancies [0.505(4) vs.
3.1. Synthesis and basic characterization
Three 1,4-diphenylpiperazine-2,5-diones, with -H (1), -NO2 (2)
and -N(CH3)2 (3) substituents on the para- position of phenyl rings
were prepared by cyclization of two chloroacetamide molecules in
acetonitrile, in the presence of NaOH as a strong base (Scheme 1)
[26]. The chloroacetamide precursors (P1-P3) for the synthesis of
1–3 were obtained by chloroacetylation of aniline, 4-nitroaniline
and 4-(dimethylamino)aniline, respectively (see Supplementary
material). The products 1–3, obtained in yields 43–66%, were char-
acterized by 1H and 13C NMR spectroscopy in solution, IR(KBr)
spectroscopy and ESI-MS and HRMS spectrometry.
0
.495(4) for A and B parts, respectively] were obtained. All experi-
mental and refinement data for the single crystal X-ray diffraction
studies in this publication are collected in Table 1.
CCDC 2043246–2043248 contains the supplementary crystal-
lographic data for this paper. These data can be obtained free
of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html (or
from the Cambridge Crystallographic Data center, 12, Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336,033).
When comparing the IR spectra of P1-P3 to 1–3, chloroac-
etamides P1-P3 have peaks of large intensity around 3300–3000
Table 1
cm 1, possibly corresponding to the secondary amine N–H stretch-
ing. The C=O band moved to lower wavenumbers in 1 and 2 and to
higher wavenumbers in 3, compared to P1, P2 and P3, respectively.
Unlike the solid-state spectra, in solution state the NMR spectra in-
dicate symmetry.
−
Single crystal X-ray diffraction data.
1
2
3
Formula
Mr (g mol-1
C16H14N2O2
266.29
C16H12N4O6
356.30
C20H24N4O2
352.43
)
Crystal size (mm)
Crystal color
0.2 × 0.15 × 0.1 0.3 × 0.2 × 0.1 0.2 × 0.15 × 0.1
colorless
monoclinic
P21
light yellow
orthorhombic
Pbca
light brown
monoclinic
P21/c
Crystal system
3
.2. ESI-MS and HRMS spectrometry
Space group
a ( A˚ )
b ( A˚ )
c ( A˚ )
α (o)
β (o)
γ (o)
5.9159(5)
7.7117(9)
14.2562(13)
90
14.7583(7)
6.7230(3)
15.3267(7)
90
8.8426(4)
6.2859(3)
16.4710(7)
90
The synthesized compounds were identified by two mass spec-
trometric methods (ESI and MALDI).
The ESI and MALDI mass spectra of 1,4-diphenylpiperazine-
99.719(9)
90
90
90.997(4)
90
2
,5–dione (1) are shown in Figures S7a-d. In the ESI mass spec-
90
trum measured in the positive mode the signal of the protonated
V ( A˚ 3
)
641.06(11)
2
1520.72(12)
4
915.38(7)
2
+
+
molecular ion, [M+H] ([C16 H14 N O +H] ) at m/z 267.2 (calcd
Z
2
2
Z’
1
1/2
1/2
266.1055) is the dominant peak. Similarly, in the positive mode
of the MALDI MS the signal at m/z 267.1120 is also detected.
These results undoubtedly confirm the identity of the cation of 1,4-
diphenylpiperazine-2,5–dione.
Dcalc
1.380
1.556
1.279
F(000)
280
736
376
Reflns. collected
Independ. reflns.
Rinit
2237
6932
2663
1811
1570
1402
The ESI and MALDI mass spectra of the 1,4-bis(4-
nitrophenyl)piperazine-2,5–dione (2) are given in Figures S8a-d.
Due to low solubility of 2, the ESI mass spectra could not be in-
terpreted with certainty. In the positive mode of the MALDI mass
0.0203
1602
0.0810
1357
0.0180
1066
Reflns. observed
Parameters
R[I>2σ(I)]a
wR2 (all data)b
Goof, Sc
181
118
148
0.0651
0.1771
1.050
0.0762
0.2152
1.068
0.0412
0.1153
1.048
+
+
spectrum (Figure S8d), the signal [M+H] ([C16 H12N4O6+H] )
at m/z 357.0830 (calcd 356.0757) is detected. However, this
protonated molecular ion is not the dominant ion in the spectrum.
In the positive mode of the ESI and MALDI mass spectra
of 1,4-bis(4(dimethylamino)phenyl)-piperazine-2,5–dione (3), sig-
nals are detected that correspond to its protonated molecular ion
Max./Min. res. el. dens. +0.389/–0.286
+0.450/–0.304 +0.102/–0.124
a
R = ꢁ||Fo|–|Fc||/ ꢁ|Fo|.
b
c
wR2 = {ꢁ[w(Fo2 – Fc2)2]/ ꢁ[w(Fo2)2]}1/2
.
S = {ꢁ[w(Fo – Fc2)2]/(n–p)}1/2 where n is the number of reflections and p is
2
the total number of parameters refined.
3

B. Peri ´c , N.P. Juraj, G. Szalontai et al.
Journal of Molecular Structure 1234 (2021) 130157
+
+
[
M+H] ([C20H24N O +H] ) at m/z 353.3 (ESI, Figure S9a) and m/z
Hꢀꢀꢀπ interactions between phenyls, these should be the main in-
termolecular interactions governing the crystal packing arrange-
ments of molecules in solid-state structures of 1, 2 and 3. Com-
plete analyses of these intermolecular interactions are given in the
Supplementary material (Figures S19-S27 and Tables S5-S11).
For 1, several C–HꢀꢀꢀO interactions connect molecules along
crystallographic axis a (Figure S19). Although piperazines are not
π conjugated systems, they are planar in all three structures 1, 2,
and 3. In the case of structure 1, they are almost parallel with one
of the phenyl rings from neighboring molecules (Ph2), with a di-
hedral angle between the piperazine ring plane and nearest phenyl
rings plane of only 1.10(19)o [1°]. These interactions, supported
4
2
3
53.1964 (MALDI, Figure S9d).
3
.3. Single crystal X-ray and DFT structures
X-ray single crystal structures were determined for all three
compounds in this study (1, 2 and 3). For 3, a disorder was present,
due to the different orientation of the central piperazine ring.
Their molecular structures, with atomic numbering schemes are
given in Figures S16-S18 of Supplementary material. All structures
were also optimized by pseudopotential plane-wave DFT calcula-
tions in periodic potential, in order to obtain reliable NMR param-
eters [12,16]. In the analysis given below, all values for distances
and angles given without parentheses are from the X-ray diffrac-
tion structures and those given in parentheses are from the DFT
optimized structures, (valid also for Tables S5-S11 of Supplemen-
tary material). The conformations of the central piperazine ring in
iii
by one additional C–HꢀꢀꢀO interaction (C10–H10AꢀꢀꢀO2 ) and one
iii
C–Hꢀꢀꢀπ interaction (C10–H10AꢀꢀꢀPh2 ) connect molecules along
crystallographic axis b (Figure S20). In this way two-dimensional
assemblies of molecules are formed parallel with the ab planes of
the unit cell. Along the axis c, molecules (i.e. the assemblies) are
connected only by weak C–Hꢀꢀꢀπ interactions existing only among
1
–3 are almost planar, with good agreement of both experimental
v
SC-XRD and theoretical DFT data. The nitrogen and carbon atoms
one phenyl type (C5–H5ꢀꢀꢀPh1 , Figure S20). These phenyls do not
˚
show any interaction with piperazine rings, they are mutually in-
clined (the angle between the least square planes of neighboring
phenyl rings Ph1 and Ph1v rings is 64.3(2)o [64°]). According to the
search criteria of program PLATON [29] only in the DFT optimized
model of 1, the C5–H5ꢀꢀꢀPh1v interaction is detected (Table S7 in
Supplementary material), witnessing the weakness of Ph1ꢀꢀꢀPh1
intermolecular interactions compared with other intermolecular
interactions for phenyl ring Ph2. It can be said that this difference
in intermolecular interactions for two phenyl rings of one molecule
of 1 is the reason for their crystallographic (i.e. symmetry) inequiv-
alence. This symmetry breaking argument is opposite from the
symmetry breaking argument given in the analysis of solid-state
structures of two 9,10-anthracene amino acid conjugates: Ant-(CO-
Phe-OMe)2, and Ant-(CO-Gly-Phe-OMe)2 [15]. In those anthracene
structures, the origin of symmetry breaking was intramolecular,
i.e. the chirality of amino-acids, while intermolecular interactions
had almost centrosymmetric features [15]. Here, the molecules of
1 have no chiral atoms at all, but intermolecular interactions break
the symmetry into the non-centrosymmetric space group P21. The
main crystal packing feature of compound 1 is illustrated in the
left part of Fig. 2a, where two-dimensional supramolecular aggre-
gates parallel with the ab planes of the unit are emphasized by red
color and bounded between two red-colored planes. Phenyl rings
Ph2 and piperazine rings form these aggregates, while phenyl rings
Ph1 are outside.
forming the piperazine ring are within 0.064 [0.036] A apart from
the least-square plane which they define. Carbonyl oxygen atoms,
as parts of piperazine groups, are within these planes: the largest
˚
deviation is 0.103(3) [0.105] A. It is important to notice that crys-
tallographic centers of inversion exist at the centers of piperazine
rings in structures 2 and 3, but not in structure 1, although the
molecular conformation of 1 is almost centrosymmetric (Figure S16
in Supplementary material). This results in a crystallographic (sym-
metrical) independency for all atoms in one molecule of 1 (two
carbonyl carbon atoms C7 and C9, two ipso- carbon atoms C1 and
C11, etc. …), while for 2 and 3, the number of symmetry inde-
pendent atoms is reduced to one half of the molecule (Z’ = 1/2,
Table 1). However, for 3, two possible conformations of the piper-
azine ring increases the number of symmetry independent posi-
tions atoms can occupy.
Due to the observed crystallographic symmetry breaking in 1,
there are two characteristic angles between planes of piperazine
and phenyl rings: 58.0(2)o [57.19°] for one side of the molecule and
2.55(19)o [51.55°] for the other side. In 2, this angle is equal on
both sides of the molecule and it is significantly larger, 78.61(10)o
5
[
75.01°]. For one piperazine orientation in 3 (A disorder part) this
angle is 73.4(5)o [70.24°], similar to 2, and for the other piperazine
orientation (B disorder part) this angle is 52.2(4)o [56.73°] similar
to 1.
o
For 2, the nitro group substituent (NO ) is inclined by 19.8(3)
2
[
15.90°] with respect to the phenyl ring, with para- carbon atom
Crystal packing features of 2 (Fig. 2b, left part) are governed
by C–HꢀꢀꢀO interactions and πꢀꢀꢀπ stacking interactions between
phenyl rings (Figures 23–25 and Tables S8-S9), realized in the or-
thorhombic and centrosymmetric space group Pbca. The increased
number of oxygen atoms, due to presence of the nitro group, re-
sult in a higher number of C–HꢀꢀꢀO interactions. It is interesting
that all hydrogen atoms in the structure participate in one type
of C–HꢀꢀꢀO interaction. Again, this observation is realized only in
the DFT optimized structure of 2, due to slightly improved posi-
tions of hydrogen atoms with respect to automatically generated
“riding” positions obtained from X-ray diffraction analysis. Besides
the C–HꢀꢀꢀO interactions, phenyl rings show stacking interactions
along the crystallographic axis b (Figure S24). An interesting fea-
ture of the C–HꢀꢀꢀO interactions in 2 is the involvement of ortho-
hydrogen atoms of phenyl rings with oxygen atoms from the ni-
tro group. As result, nitro groups from the neighboring molecules
are positioned nearby the piperazine ring of the central molecule
(Figure S23).
C6 almost in the same plane as the nitro group [deviation is
˚ ˚
.017(2) A [0.062 A]. On the other hand, the non-hydrogen atoms
0
of N(CH3)2 group in 3 do not form a plane with the para- car-
bon atom of the phenyl ring (C6). Deviation of the nitrogen atom
˚
˚
is 0.188(2) A [0.192 A] from the plane defined by nearby bonding
carbon atoms, resulting in a conformation where methyl carbons
C1N and C2N deviate from the phenyl plane, lying on the same
side of the phenyl ring [torsion angles C1N–N2N–C6-C4 and C2N–
o
o
N2N–C6–C5 are −9.9(3) [−10.61°] and 18.1(3) [15.71°], respec-
tively]. In other words, the geometry around the nitrogen atom of
the N(CH3)2 group should be more precisely described as trigonal
pyramidal (i.e. partially tetrahedral) rather than planar, as already
discussed by Dahl [27]. However, due to relatively large atomic
displacement parameters (ADPs) for methyl carbon atoms, with
the largest components oriented perpendicular to the phenyl plane
(
Figure S18), it could be expected that these atoms exhibit motions
perpendicular the phenyl plane, even in the solid state [28].
In the analysis of intermolecular interactions, classic hydrogen
bonds with electronegative donors in O–H and N–H groups are
not present; the only type of interactions should be weaker C–
HꢀꢀꢀO interactions. Together with aromatic πꢀꢀꢀπ (stacking) or C–
The number of C–HꢀꢀꢀO interactions, or C–Hꢀꢀꢀπ interactions
are highly reduced in the crystal packing of 3 (Figures S26, S27
and Tables S10, S11). Only one acceptor, the carbonyl oxygen, is
involved in C–HꢀꢀꢀO interactions in structural models of both dis-
4

B. Peri ´c , N.P. Juraj, G. Szalontai et al.
Journal of Molecular Structure 1234 (2021) 130157
Fig. 2. Basic structural motifs (left) and 13 C-CPMAS/CPNQS spectra of compounds 1 (a), 2 (b) and 3 (c) (right). In all spectra vertical bars represent 13 C δcalc values based
on the GIPAW calculations and Eq. (1) with A = −0,970(7) and B = 167.8(5). In Figure (c) atoms, labels and δcalc values for disorder component 3A are shown black, while
atoms, labels and δcalc values for disorder component 3B are shown gray. Conditions for the measurements: 9.4 T, rotation 12 kHz, 4 mm ZrO2 rotor, room temperature.
order components: in component 3A it connects molecules along
the crystallographic axis b via a C2–H2ꢀꢀꢀO1Aii interaction (Fig-
ure S26a), and in component 3B it connects four neighboring
molecules via C5–H5ꢀꢀꢀO1Biii and C2N–H2N1ꢀꢀꢀO1Biii interactions
in the model of component 3B, found by geometry search crite-
ria, it might be concluded that this structural model is more sta-
ble than the 3A model. However, two carbonyl oxygen atoms from
neighboring molecules in model 3B are in close contact; the dis-
tance between atoms O1B and O1Biv [Symmetry code (iv): 1-x, 1-y,
(
Figure S26b). Due to a higher number of C–HꢀꢀꢀO interactions
5

B. Peri ´c , N.P. Juraj, G. Szalontai et al.
Journal of Molecular Structure 1234 (2021) 130157
Fig. 3. 15 N-CPMAS spectra of compound 1, with 15 N δcalc values based on the GIPAW calculations from this work and Eq. (1) with A = −1.004 and B = −161.5 (blue),
31
in
comparison with experimental values (black).
˚
z for 3, space group P2 /c] is only 2.500(5) [3.152] A, which is
1
ical approach [30]. The calculations were based on DFT optimized
-
rather short for the two most electronegative atoms in the struc-
ture (Figure S26b). Indeed, the DFT optimization of model 3B re-
sulted in final energy of 0.016 Ry (2.16 kJ/mol) above the energy
of the DFT optimized model 3A and significant elongation of the
O1BꢀꢀꢀO1Biv contact with respect to positions obtained from the
X-ray data. Analogous close contacts between O1A oxygen atoms
are not present in the 3A structural model, where distances be-
tween O1A and O1Aiv atoms are above 5 A. The most significant
intermolecular interactions in the structure of 3 (in both disorder
models) are two C–Hꢀꢀꢀπ interactions (C3–H3ꢀꢀꢀPh2vi and C1N–
H1N1ꢀꢀꢀPhvii. Figure S27 and Table S11), which connect the cen-
tral molecule with eight neighboring molecules, forming the main
solid-state structural motif (left part of Fig. 2c). In the crystal pack-
ing analysis of compound 3, we have to bear in mind that the oc-
cupancies of O1A and O1B are 0.505(4) and 0.495(4), respectively,
thus the number of interactions involving O1A or O1B atoms are
reduced by the same values. Also, one orientation of the piperazine
ring in the central molecule does not imply the same orientation
of piperazine ring in neighboring molecules, although such orien-
tations are automatically conserved in both proposed models 3A
structures of 1 and 2 and on both DFT optimized models for com-
pound 3 (3A and 3B), described in the previous chapter. Results
of calculations are given in Tables S12-S14 and S16-S18 of Supple-
mentary material. Basic structural motifs for compounds 1–3, with
labelled individual carbon and nitrogen sites are given on the right
parts of Figs. 2a-c.
Calculated NMR shieldings (σ) can be compared with exper-
imental chemical shifts (δexp) through calculated chemical shifts
(δcalc) obtained from σ by a linear equation:
˚
δcalc = A × σ + B
(1)
Parameters A (coeff.) and B (intercept) should be universal for
particular nuclei type and for particular theoretical model (calcula-
tion procedure, DFT functional, etc.). They can be used from the lit-
erature or could be determined by linear regression analysis from
the experimental data set of the assigned δexp signals. For purpose
of assignation from the spectra, A and B parameters proposed by
M. Marín-Luna et al. [31] are used, (A = –0.953 and B = 165.1,
parameters proposed for GIPAW calculations). According to these
predictions (Figures S31-S36 of Supplementary material), the ma-
jority of the signals in the experimental spectra were assigned. In
some cases, two carbon sites are assigned to one experimentally
observed signal (C1 and C11 in 1, C2 and C3, C4 and C5 for 2, etc.),
for methyl carbon atoms in the spectra of 3, one signal is assigned
to both methyl carbon atoms of model 3A and both methyl car-
bon atoms of model 3B (four sites). Due to the large number of
symmetry independent aromatic C–H carbon atoms in compounds
and 3B, by keeping the P2 /c space group symmetry. Thus, addi-
1
tional models for the structure of compound 3 are possible, which
should be more correctly described in the lower space group sym-
metries, possibly with increased volume of the unit cell. Such more
detailed models of disorder observed in structure of 3 are not ac-
counted in this work because a satisfactory electron density map,
in accordance with X-ray diffraction intensities, is achieved.
1
and 3, the experimentally found broad peaks in the correspond-
3
.4. Solid-state NMR measurements and GIPAW calculations
ing regions of 13C CPMAS spectra (Figures S31 and S35) are not
assigned to the particular atoms. Although predictions based on
parameters A and B given by M. Marín-Luna et al. [31] agree re-
markably well with the observed signals, there are significant dif-
ferences for amide carbon atoms for all three compounds (Figures
S31-S36). For this reason a linear regression analysis between ex-
perimentally determined 13C shifts (δexp) and calculated shieldings
(σ) for 26 assigned carbon sites of 1, 2 and 3 is performed. Details
are given in the Supplementary material [R2 values, standard de-
viations for fit and for individual parameters A and B of Eq. (1)].
Graphical presentation of the linear correlation equation is given
in Figure S37 of Supplementary material. Redefined parameters for
To study the influence of structural features of compounds 1,
and 3 on NMR properties, 13C-CPMAS and 13C-CPNQS solid-
2
state NMR measurements for all compounds were performed (right
parts of the Figs. 2a-c), as well as 15N-CPMAS solid-state NMR
measurements for compound 1 (Fig. 3). Individual spectra are also
given in Figures S31-S36, S38 of Supplementary material, which
contain the experimentally determined values of signals. At the
same time, the NMR shieldings and quadrupolar interaction pa-
rameters (quadrupolar coupling constants, CQ and electric field
gradients) for all atoms were calculated using the GIPAW theoret-
6

B. Peri ´c , N.P. Juraj, G. Szalontai et al.
Journal of Molecular Structure 1234 (2021) 130157
equation (1) are: A = -0,970(7) and B = 167.8(5). Vertical bars in
calc
NMR spectra analogous with compound 2 for which also Z’ = 1/2
(Table 1). Therefore, only one amide signal and only one ipso- sig-
nal are expected for each of the individual structural models. Only
superposition of these models (disorder) can produce the observed
doubling of signals. This type of doubling has a different origin
than the “symmetry breaking” argument encountered in 1 or in
13C ssNMR spectra of 9,10-anthracene amino acid conjugates ana-
lyzed in our previous work [15]. In 1, every atom of one molecule
has a different structural environment due to specific intermolec-
ular interactions. Piperazine ring in 3 has two different orienta-
tions, i.e. there exist two molecular conformations for molecules of
compound 3 in the structure. Molecules of each conformation are
randomly distributed across the crystallites in the sample. As al-
ready mentioned in analysis of the crystal structure (chapter 3.3),
additional more detailed models for disorder of compound 3 re-
quire models with several molecules in the asymmetric unit (Z’>1),
possibly with increased unit cell dimensions. Regarding solid-state
NMR measurements in this work, such additional models are not
necessary because two-component models 3A and 3B are enough
to simulate the experimental doubling of amide and ipso- carbon
13C NMR signals in the spectra.
the left parts of Figs. 2a-c are positioned on the values of δ
cal-
culated by such redefined parameters A and B, for all carbon sites
in the structures.
Although differences between δcalc and δexp for amide carbon
sites are reduced, with respect to those calculated from A and B
parameters of M. Marín-Luna et al. [31], δexp are still positioned
downfield with respect to δcalc. On the other hand, aromatic C–H
carbon sites for 2 (C2/C3 and C4/C5) are experimentally positioned
upfield with respect to δcalc (Fig. 2b, left part). It is interesting to
note that predictions for these atoms based on the parameters of
M. Marín-Luna et al. [31] correspond to experimental peaks very
well (Figure S33). The largest difference of 4 ppm is observed for
quaternary aromatic carbon site C1 in the structure of 2, although
assignation of this site is undoubted, because it appears in both
types of the spectra.
For 1, due to symmetry independence of all atoms in the
molecule (Z’ = 1), two closely predicted peaks for all chemically
identical atom sites are expected. However, overlaps into single ob-
servable signal are possible. Two signals for amide carbon atoms
C7 and C9 are only barely visible in 13C-CPMAS spectra, while in
the 13C-CPNQS spectra they are overlapped into one peak possess-
ing enlarged linewidth (~ 250 Hz, right part of Fig. 2a). Quaternary
ipso- carbons C1 and C11 are overlapped in both spectra, with re-
duced linewidth, in accordance with a smaller difference of their
predicted shifts values. A good prediction for two signals of two
symmetry independent CH2 carbon atoms of the piperazine group
However, the question is open about observation of the single
signals for para- carbons C6, methyl carbons C1N and C2N and for
CH2 carbons C8A and C8B, although superposition of two mod-
els 3A and 3B produce two distinct δcalc predictions (Fig. 2c). This
question is especially addressed to the carbon atoms C8A and C8B
because they are parts of the disordered piperazine ring. This effect
could be connected with mobilities of atoms in solid state struc-
tures, visible through the atomic displacement parameters (ADP) of
atoms obtained from the SC-XRD [28]. Additionally, all analogously
labeled atoms outside the piperazine rings have similar positions
in both DFT optimized structures (Fig. 2c, left part), thus ordinary
vibrational motions in the solid state can exchange these positions.
Increased ADP parameters and similar positions in both disordered
models are observed for para- and methyl carbon atoms which
are far away from the disordered piperazine ring. Internal rota-
tions around the N–Cpara bond at room temperature also cannot
be excluded. Such rotations may be rapid compared to the diffrac-
tion recording time but would not be apparent from the crystal
structure if it occurs with a jump motion [33]. Therefore, by the
inclusion of the rotational/vibrational corrections in the chemical
shielding (shift) calculation procedures, [34] a better correspon-
dence between predicted positions for individual sites and experi-
mentally observed signals could be achieved.
(
C8 and C10) is clearly visible in the right part of Fig. 2a).
Further evidence of such “doubling” of signals for chemically
equivalent atom sites in compound 1 is also obtained by the 15N-
CPMAS NMR measurements (Fig. 3). Two observed 15N chemi-
cal shifts (experimental values are given in Figure S38 in Sup-
plementary material) are in very good agreement with our GI-
PAW calculations (values for N1 and N2 in Table S12) and corre-
lation Eq. (1) with parameters A and B proposed by M. Marín-Luna
et al. [31], [values suggested for the GIPAW calculated 15N shifts:
A = −1.004 and B = −161.5].
For 2 (Fig. 2b), only six (three) different signals are observed
in corresponding 13C-CPMAS (13C-CPNQS) spectra, in accordance
with the crystallographic structure of centrosymmetric molecules
(Z’ = 1/2). In general, the linewidths of 2 are narrower than those
of 1 (150 Hz vs 250 Hz). The C8 signal is a single peak, its broaden-
ing could be attributed to the residual dipolar effect of the directly
attached 14 N spins [32]. Aromatic C–H carbons, ortho- (C2/C3) and
meta- (C4/C5) are two single peaks in the 13C-CPMAS spectrum; al-
though slightly shifted upfield with respect to predicted positions,
the difference between predicted positions and difference between
experimental peaks agree very well.
4. Conclusion
In conclusion, diketopiperazines 1, 2 and 3 were prepared
and their single-crystal structures determined, showing symme-
try breaking in 1, a symmetric structure in 2 and a disorder in 3.
Crystallographic structures of the corresponding compounds corre-
spond very well to the ssNMR measurements.
For 3 (Fig. 2c), two amide signals are visible in both types of
experimental spectra, two ipso- carbon signals are also visible in
the 13C-CPNQS spectrum which are barely visible in the 13C-CPMAS
spectrum (due to the overlap with signals from the C–H aromatic
atoms C2 and C3, for each disorder component), while single peaks
are observed for para- carbon atoms (C6), CH2 carbon atoms (C8A,
C8B) and both methyl carbon atoms (C1N, C2N). Signals originat-
ing from aromatic C–H carbon atoms are grouped into two wide
regions (~110–118 ppm and ~125–132 ppm) which clearly could be
In particular, the slightly asymmetrical molecule 1 produces ei-
ther signal pairs for chemically equivalent atom sites or signals
with enlarged line-widths in ssNMR, if the signals in the pair
have very close shifts. Structurally, we showed that the origin of
this molecular asymmetry is in the different crystal packing roles
of individual parts of the molecule. For this structure, one DFT
optimization and one calculation of NMR parameters in the ap-
propriate low symmetry produce satisfactory explanation of the
13C and 15N ssNMR spectra. The most symmetrical compound 2
showed the smallest number of signals and easiest assignation of
the 13C NMR signals. Molecules of 3, although symmetric, crys-
tallize with two orientations of the piperazine group, revealed as
disorder through the appearance of atoms with lowered crystal-
lographic occupancies. In this work we showed that such disor-
associated as “ortho-“ and meta-“ regions, due to the δ
values
calc
for C2/C3 and C4/C5 sites. A signal at ~124 ppm in the 13C-CPMAS
spectra (Figure S35) can be tentatively associated with one of these
aromatic C–H carbon atoms, i.e. to the C3 atom of the 3B disorder
component.
Individual structural models 3A and 3B are both based on the
symmetry of space group P2 /c, with only half of a molecule in
1
one asymmetric unit, i.e. with Z’ = 1/2, so each of these mod-
els should produce a reduced number of carbon signals in the ss-
7

B. Peri ´c , N.P. Juraj, G. Szalontai et al.
Journal of Molecular Structure 1234 (2021) 130157
der produces similar effects in ssNMR spectra as the asymmetry of
molecules in 1, i.e. the occurrence of signal pairs for the chemically
equivalent atom sites. We showed that superposition of two DFT
optimized structural models for compound 3 adequately describes
main features in the ssNMR spectra as well as observed electron
density map obtained from SC-XRD. An additional observation that
some of the signals in ssNMR spectra still appear as single peaks,
was attributed to possible rotational/vibrational motions of atom
sites, concluding that such corrections should be incorporated in
the procedures of calculation of the NMR properties.
[6] Y. Liang, P. Zhang, J. Chen, Function-oriented design of conjugated carbonyl
compound electrodes for high energy lithium batteries, Chem. Sci. 4 (2013)
1330–1337 https://doi.org/10.1039/c3sc22093a.
[
7] F. Cumine, S. Zhou, T. Tuttle, J.A. Murphy, A study of diketopiperazines as elec-
tron-donor initiators in transition metal-free haloarene–arene coupling, Org.
Biomol. Chem. 15 (2017) 3324–3336 https://doi.org/10.1039/C7OB00036G.
8] A. Maity, A. Hazra, P. Palit, S. Mondal, S. Lala, N.B. Mondal, The cy-
totoxic effects of diketopiperaizes against Leishmania donovani pro-
mastigotes and amastigotes, Med. Chem. Res. 22 (2013) 3452–3458
https://doi.org/10.1007/s00044-012-0355-9.
[
[
9] F.L. Bettens, R.P.A. Bettens, R.D. Brown, P.D. Godfrey, The microwave spectrum,
structure, and ring-puckering of the cyclic dipeptide diketopiperazine, J. Am.
Chem. Soc. 22 (2000) 5856–5860 http://doi.org/10.1021/ja000042n.
The information obtained herein can be used for the treatment
of molecules of different symmetry. Understanding the solid-state
properties of piperazine molecules is useful as diketopiperazines
are often found in biologically active compounds.
[10] D. Mukhopadhyay, C. Gupta, T. Theint, C.P. Jaroniec, Peptide bond confor-
mation in peptides and proteins probed by dipolar coupling-chemical shift
tensor correlation solid-state NMR, J. Magn. Reson. 297 (2018) 152–160
https://doi.org/10.1016/j.jmr.2018.10.015.
[11] P.A. Wabnitz, R.J. Waugh, M.A. Eckersley, S. Dua, T. Blumenthal,
J.H. Bowie, The negative ion mass spectra of deprotonated 2,5-dike-
topiperazines, Int. J. Mass Spectrom. Ion Process. 154 (1996) 193–201
https://doi.org/10.1016/0168-1176(96)04378-9.
Declaration of Competing Interest
[
12] P.
Prog.
http://doi.org/10.1016/j.pnmrs.2020.03.001.
Hodgkinson,
NMR
crystallography
of
molecular
(2020)
organics,
10–53
Nucl. Magn.
Reson. Spectrosc.
118-119
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared to
influence the work reported in this paper.
[13] D.L. Bryce, NMR crystallography: structure and properties of materials from
solid-state nuclear magnetic resonance observables, IUCrJ 4 (2017) 350–359
http://doi.org/10.1107/S2052252517006042.
[14] I.J. King, F. Fayon, D. Massiot, R.K. Harris, J.S.O. Evans, A space group assign-
ment of ZrP2O7 obtained by 31 P solid state NMR, Chem. Commun. (2001)
CRediT authorship contribution statement
1766–1767 https://doi.org/10.1039/b106379k.
[
15] B. Peri c´ , G. Szalontai, M. Borovina, D. Viki c´ -Topi c´ , S.I. Kirin, Symmetry break-
ing in the solid state of 9,10-anthracene amino acid conjugates as seen by
X-ray diffraction and NMR spectroscopy, J. Mol. Struc. 1221 (2020) 128834
https://doi.org/10.1016/j.molstruc.2020.128834.
Berislav Peri c´ : Conceptualization, Formal analysis, Methodol-
ogy, Writing - original draft. Natalija Pantalon Juraj: Formal anal-
ysis, Resources, Writing - review & editing. Gábor Szalontai:
Investigation, Resources, Writing - review & editing. Suzana R.
Veli cˇ kovi c´ : Investigation, Writing - review & editing. Filip M.
[
16] S.E. Ashbrook, D. McKay, Combining solid-state NMR spectroscopy with
first-principles calculations – a guide to NMR crystallography, Chem. Commun.
52 (2016) 7186–7204 https://doi.org/10.1039/c6cc02542k.
[
17] N. P. Juraj, J. Le Pennec, B. Peri c´ , S.I. Kirin, A Case study of supramolecular
organization: one ferrocene substituted iminodiacetamide and its chloroform
solvate, Croat. Chem. Acta 90 (2018) 613–623 https://doi.org/10.5562/cca3270.
[18] N. Pantalon Juraj, G. Mileti c´ , B. Peri c´ , Z. Popovi c´ , N. Smre cˇ ki, R. Vianello,
S.I. Kirin, Stereochemistry of hexacoordinated Zn(II), Cu(II), Ni(II), and
Co(II) complexes with iminodiacetamide ligands, Inorg. Chem 58 (2019)
Veljkovi c´ : Investigation, Writing
- review & editing. Dražen
Viki c´ -Topi c´ : Investigation, Writing - review & editing. Sre c´ ko I.
Kirin: Conceptualization, Funding acquisition, Project administra-
tion, Writing - review & editing.
1
6445–16457 https://doi.org/10.1021/acs.inorgchem.9b02200.
[19] Agilent Technologies, CrysAlis Pro Software, Yarnton, Oxfordshire, England,
012 Ver. 1.171.39.46.
20] G.M. Sheldrick,
tal-structure determination,
https://doi.org/10.1107/S2053273314026370.
21] G.M. Sheldrick, Crystal structure refinement with SHELXL, Acta Cryst C71
2015) 3–8 https://doi.org/10.1107/S2053229614024218.
Acknowledgments
2
[
[
SHELXT
-
Integrated
Acta Cryst.
space-group
A71
and
(2015)
crys-
3–8
The presented materials are based on work financed by the
Croatian Science Foundation through the research project “Min-
imal artificial enzymes” (HrZZ, IP-2014–09–1461) and in part by
(
a
Croatian-Serbian bilateral project. The authors thank Ernest
[
22] P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni,
D. Ceresoli, G.L. Chiarotti, M. Cococcioni, I. Dabo, A. Dal Corso, S. Fab-
ris, G. Fratesi, S. de Gironcoli, R. Gebauer, U. Gerstmann, C. Gougous-
sis, A. Kokalj, M. Lazzeri, L. Martin-Samos, N. Marzari, F. Mauri, R. Maz-
zarello, S. Paolini, A. Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo,
G. Sclauzero, A.P. Seitsonen, A. Smogunov, P. Umari, R.M. Wentzcovitch, QUAN-
Sanders (RBI) for technical assistance. The computations were per-
formed using the resources of the computer cluster Isabella based
in the University Computing center SRCE - University of Zagreb.
TUM ESPRESSO:
a modular and open-source software project for quan-
Supplementary materials
tum simulations of materials, J. Phys.: Condens. Matter 21 (2009) 395502
https://doi.org/10.1088/0953-8984/21/39/395502; (b) P. Giannozzi, O. An-
dreussi, T. Brumme, O. Bunau, M. Buongiorno Nardelli, M. Calandra, R. Car,
C. Cavazzoni, D. Ceresoli, M. Cococcioni, N. Colonna, I. Carnimeo, A. Dal Corso,
S. de Gironcoli, P. Delugas, R.A. DiStasio Jr, A. Ferretti, A. Floris, G. Fratesi,
G. Fugallo, R. Gebauer, U. Gerstmann, F. Giustino, T. Gorni, J. Jia, M. Kawa-
mura, H.-.Y. Ko, A. Kokalj, E. Küçükbenli, M. Lazzeri, M. Marsili, N. Marzari,
F. Mauri, N.L. Nguyen, H.-.V. Nguyen, A. Otero-de-la-Roza, L. Paulatto, S. Poncé,
D. Rocca, R. Sabatini, B. Santra, M. Schlipf, A.P. Seitsonen, A. Smogunov, I. Tim-
rov, T. Thonhauser, P. Umari, N. Vast, X. Wu, S. Baroni, Advanced capabilities
for materials modelling with quantum ESPRESSO, J. Phys.: Condens. Matter 29
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.molstruc.2021.130157.
References
[
1] A.D. Borthwick, 2,5-diketopiperazines: synthesis, reactions, medicinal chem-
istry, and bioactive natural products, Chem. Rev. 112 (2012) 3641–3716
https://doi.org/10.1021/cr200398y.
(2017) 465901 https://doi.org/10.1088/1361-648X/aa8f79.
[
2] M.B. Martins, C. Ivone, Diketopiperazines: biological activity and synthesis,
Tetrahedron 63 (2007) 9923–9932 https://doi.org/10.1016/j.tet.2007.04.105.
3] K.C. Nicolaou, M. Lu, S. Totokotsopoulos, P. Heretsch, D. Giguère, Y.-.P. Sun,
D. Sarlah, T.H. Nguyen, I.C. Wolf, D.F. Smee, C.W. Day, S. Bopp, E.A. Winzeler,
Synthesis and Biological evaluation of epidithio-, epitetrathio-, and
bis-(methylthio)diketopiperazines: synthetic methodology, enantioselec-
[
23] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient ap-
proximation made simple, Phys. Rev. Lett. 77 (1996) 3865–3868
https://doi.org/10.1103/PhysRevLett.77.3865.
[
[
24] https://sites.google.com/site/dceresoli/pseudopotentials. (accessed
021).
5 January
2
ꢀ
[25] N. Varini, D. Ceresoli, L. Martin-Samos, I. Girotto, C. Cavazzoni, Enhancement
of DFT-calculations at petascale: nuclear magnetic resonance, hybrid density
functional theory and car–parrinello calculations, Comp. Phys. Comm. 184
tive total synthesis of epicoccin G, 8,8 -epi-ent-rostratin b, gliotoxin,
gliotoxin g, emethallicin e, and haematocin and discovery of new antivi-
ral and antimalarial agents, J. Am. Chem. Soc. 134 (2012) 17320–17332
https://doi.org/10.1021/ja308429f..
(2013) 1827–1833 https://doi.org/10.1016/j.cpc.2013.03.003.
[
26] S.-.D. Cho, S.Y. Song, K.-.H. Kim, B.-.X. Zhao, C. Ahn, W.H. Joo, Y.-J. Yoon,
J. Falck, D.-S. Shin, One-pot synthesis of symmetrical 1,4-disubstituted
piperazine-2,5-diones, Bull. Korean Chem. Soc. 25 (3) (2004) 415–416
https://doi.org/10.5012/bkcs.2004.25.3.415.
[
4] D.L. Boger, B.E. Fink, M.P. Hedrick, A new class of highly cytotoxic
diketopiperazines, Bioorg. Med. Chem. Lett. 10 (2000) 1019–1020
https://doi.org/10.1016/S0960-894X(00)00152-9.
[
5] J. Geng, J.-P. Bonnet, S. Renault, F. Dolhem, P. Poizot, Evaluation of polyketones
with N-cyclic structure as electrode material for electrochemical energy stor-
age: case of tetraketopiperazine unit, Energy Environ. Sci. 3 (2010) 1929–1933
http://doi.org/10.1039/c0ee00126k.
[
27] T. Dahl, Structure from
a
twinned crystal of the triclinic form
of the 1:1 complex between N,N-dimethylaniline and hexafluo-
robenzene [C8H11 N.C6F6] at 120 K, Acta Cryst. C41 (1985) 931–933
8

B. Peri ´c , N.P. Juraj, G. Szalontai et al.
Journal of Molecular Structure 1234 (2021) 130157
https://doi.org/10.1107/S0108270185006060; (b) T. Dahl, Intermolecu-
lar interactions and -molecular geometry in molecular complexes of
N,N-di-methyl-aniline: the 1:1 complex with 1,2,4,5-tetra-cyano-benzene,
Acta Cryst C56 (2000) 708–710 https://doi.org/10.1107/S0108270100003590.
of 13 C and 15 N CPMAS NMR chemical shifts of aromatic and hetero-
cyclic bromo derivatives, Solid State Nucl. Mag. Reson. 108 (2020) 101676
https://doi.org/10.1016/j.ssnmr.2020.101676.
[32] R.K. Harris, A.C. Olivieri, Quadrupolar effects transferred to spin-12 magic-an-
gle spinning spectra of solids, Prog. Nucl. Magn. Reson. Spectrosc. 24 (1992)
435–456 https://doi.org/10.1016/0079-6565(92)80004-Y.
[
28] C.C. Wilson,
A
basic introduction to thermal motions of atoms in
crystal structures, the underlying potentials and the physical informa-
tion available from their analysis, Crystallogr. Rev. 15 (2009) 3–56
https://doi.org/10.1080/08893110802564245.
[33] R.K. Harris, P. Hodgkinson, C.J. Pickard, J.R. Yates, V. Zorin, Chemical shift com-
putations on a crystallographic basis: some reflections and comments, Magn.
Reson. Chem. 45 (2007) S174–S186 https://doi.org/10.1002/mrc.2132.
[34] T.A. Ruden, K. Ruud, Ro-Vibrational Corrections to NMR Parameters, in:
M. Kaupp, M. Bühl, V.G. Malkin (Eds.), Calculation of NMR and EPR Param-
eters, Theory and Applications, WILEY-VCH Verlag GmbH & Co. KGaA, Wein-
heim, 2004, pp. 153–173. https://doi.org/10.1002/3527601678.ch10.
[
29] A.L. Spek, Structure validation in chemical crystallography, Acta Cryst D65
(2009) 148–155 https://doi.org/10.1107/S090744490804362X.
[
30] C.J. Pickard, F. Mauri, All-electron magnetic response with pseu-
dopotentials: NMR chemical shifts, Phys. Rev.
https://doi.org/10.1103/PhysRevB.63.245101.
B
63 (2001) 245101
[
31] M. Marín-Luna, R.M. Claramunt, C. López, M. Pérez-Torralba, D. Sanz, F. Re-
viriego, I. Alkorta, José. Elguero, GIPAW versus GIAO-ZORA-SO study
A
9