Binary amorphous solids consisting of 2,4,6-triarylphenoxyl radicals and their dimers

Article abstract of DOI:10.1016/j.tetlet.2017.05.040

Although molecular amorphous materials represent an important area of research in solid-state chemistry, studies pertaining to these systems have been restricted almost exclusively to amorphous solids based on a single molecule. In this study, we found that, while the 2,4,6-bis(4-tert-butylphenyl)phenoxyl radical (2_M) and its dimer (2_D) did not give single-component amorphous solids, they rapidly formed the corresponding binary amorphous solid IIa following their condensation from benzene, dichloromethane, chloroform, and ethyl acetate solutions. The formation of IIa could be attributed to the good solubilities of 2_M and 2_D in these solvents and the high packing efficiencies of these amorphous solids. IIa was also obtained when crystals of 2_D (IIb) were ground together. The solid-state formation of IIa would not only involve the locational exchange of 2_M and 2_D, but would also involve chemical exchanges.

Full text of DOI:10.1016/j.tetlet.2017.05.040

Tetrahedron Letters 58 (2017) 2547–2550
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Binary amorphous solids consisting of 2,4,6-triarylphenoxyl radicals and
their dimers
a
a
a
a
a
Naoto Hayashi a, , Taisetsu Ueno , Naoki Okamoto , Takahiro Mori , Naho Sasaki , Taku Kamoto ,
⇑
Junro Yoshino a, Hiroyuki Higuchi a, Hidehiro Uekusa b, , Hideyuki Tukada c,
⇑
⇑
a Graduate School of Science and Engineering, University of Toyama, 3190 Gofuku, Toyama 930-8555, Japan
b Department of Chemistry and Materials Science, Tokyo Institute of Technology, Ookayama, Meguro-ku, Tokyo 152-8551, Japan
c International Graduate School of Arts and Sciences, Yokohama City University, 22-2 Seto, Kanazawa-ku, Yokohama 236-0027, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Although molecular amorphous materials represent an important area of research in solid-state chem-
istry, studies pertaining to these systems have been restricted almost exclusively to amorphous solids
based on a single molecule. In this study, we found that, while the 2,4,6-bis(4-tert-butylphenyl)phenoxyl
radical (2M) and its dimer (2D) did not give single-component amorphous solids, they rapidly formed the
corresponding binary amorphous solid IIa following their condensation from benzene, dichloromethane,
chloroform, and ethyl acetate solutions. The formation of IIa could be attributed to the good solubilities of
2M and 2D in these solvents and the high packing efficiencies of these amorphous solids. IIa was also
obtained when crystals of 2D (IIb) were ground together. The solid-state formation of IIa would not only
involve the locational exchange of 2M and 2D, but would also involve chemical exchanges.
Ó 2017 Elsevier Ltd. All rights reserved.
Received 28 March 2017
Revised 10 May 2017
Accepted 12 May 2017
Available online 13 May 2017
Keywords:
Binary amorphous
Phenoxyl
Monomer-dimer equilibrium
Chemical exchanges in solid state
Multicomponent (binary, ternary, and increasingly complex
systems) crystals represent diverse materials for research towards
solid-state chemistry and physics because they can impart various
functional characteristics, which can be used to tune the properties
of other components.1 Moreover, numerous research opportunities
still exist to expand upon our collective understanding of this area
with new features. These advantages are also promising for the
development of multicomponent amorphous solids,2 which exhibit
no long-range order between their individual components. Based
on their unique structural features, amorphous solids composed
of alloys, metal oxides, and polymers have been used extensively
as structural, magnetic, and electronic materials. Molecular amor-
phous materials have also been extensively studied. Amorphous
drugs have several advantages over the corresponding crystalline
materials in terms of their high stability and solubility.3 Further-
more, given that exciton decay processes can be avoided using
amorphous solids, amorphous thin films can be used as charge
transport layers in organic light-emitting diodes.4 Although a large
number of these amorphous solids exist as multicomponent sys-
tems, their amorphous characteristics originate from a single host
component, with the other guest components simply behaving as
passengers without disturbing the amorphous characteristics of
the host lattice. In some cases, although the individual components
of a mixture can crystallize under general conditions, as a mixture
they can readily form amorphous solids. For example, ether-
isopentane-alcohol, 3-methylpentane-isopentane, and ethanol-
methanol mixtures are frequently used to prepare frozen solutions,
which exist as amorphous solids. These amorphous solids can be
referred to as true multicomponent amorphous solids.
To further diversify the field of solid-state chemistry, it is
important to develop new true multicomponent amorphous solids.
Despite the need for further development in this area, studies per-
taining to new true multicomponent amorphous solids are rare. In
this study, we have investigated the preparation of true binary
amorphous solids consisting of 2,4,6-triarylphenoxyl radicals
(1M and 2M) and their dimers (1D and 2D).5 2,4,6-Triarylphenoxyl
radicals generally exist in a monomer-dimer equilibrium in solu-
tion.6 In contrast, however, 2,4,6-tri-tert-butylphenoxyl radicals
exist almost exclusively as monomeric radicals in the solid and
solution state because of the bulky substituents at their 4-posi-
tions.7 The condensation of solutions of almost all of the known
2,4,6-triarylphenoxyl radicals results in the formation of crystals
of the corresponding dimeric species. This is also the case for
2,4,6-triphenylphenoxyl radicals (1M), which exists in a mono-
mer-dimer (1M/1D) equilibrium in solution, as depicted in
Scheme 1. Furthermore, the condensation of solutions of 1M and
⇑
Corresponding authors.
E-mail addresses: nhayashi@sci.u-toyama.ac.jp (N. Hayashi), uekusa@chem.
titech.ac.jp (H. Uekusa), htht@yokohama-cu.ac.jp (H. Tukada).
http://dx.doi.org/10.1016/j.tetlet.2017.05.040
0040-4039/Ó 2017 Elsevier Ltd. All rights reserved.

2548
N. Hayashi et al. / Tetrahedron Letters 58 (2017) 2547–2550
Fig. 1. XRD charts (kCu, Ka = 1.5418 Å) of IIa (a) and IIb (b).
2M, the IR absorption peaks of 2M and 2D were simulated using
(U)B3LYP/6-31G(d,p) calculations and compared with the observed
IR spectra. As shown in Fig. 3 (bottom), the simulation accurately
reproduced the observed spectra of IIb. Furthermore, the differ-
ences between IIa and IIb were in good agreement with the simu-
lated peaks of 2M, indicating that 2M was indeed present in IIa.
Given that all of the peaks observed in the IR spectrum of IIa were
assigned to 2D and 2M, IIa was confirmed to consist entirely of
these molecules.
Scheme 1. Monomer-dimer equilibria of 1 M/1D and 2 M/2D.
1D leads to the formation of dimer crystals (Ib),8 irrespective of the
solvent used. In contrast, the solidification behavior of 2,4,6-tris(4-
tert-butylphenyl)phenoxyl radicals (2M) was different to that of
1M.9 While hexane solutions of 2M and 2D gave dimeric crystals
of IIb, benzene solution of these materials gave the purple glassy
solid IIa. Dimroth and co-workers referred to IIa as a ‘‘glasartige
Rückstand” (glassy residue),10 the purple color of which was attrib-
uted to the presence of 2M. There have, however, been no further
reports describing IIa in the literature.
Phenoxyl radical 2M was prepared by the oxidation of the corre-
sponding phenol with K3[Fe(CN)6] under basic conditions (See
Electronic Supplementary Information).8 The condensation of the
resulting purple solution consisting of 2M and 2D afforded the deep
purple solid IIa. The analysis of this solid by X-ray powder diffrac-
tion (XRD) revealed no diffraction peak, which indicated that IIa
was an amorphous solid (Fig. 1). It is noteworthy that the treat-
ment of 2,4,6-triphenylphenol (precursor of 1M) under the same
conditions gave virtually colorless crystals of Ib consisting exclu-
sively of 1D (Fig. 2(a)). The condensation of a dichloromethane,
chloroform, or ethyl acetate solution of 2M and 2D afforded only
IIa. In contrast, the condensation of a hexane or acetone solution
of 2M and 2D resulted in the formation of virtually colorless crystals
of IIb consisting of 2D, which was confirmed by single-crystal X-ray
analysis (Fig. 2(b)).
We initially considered the possibility that IIa consisted entirely
of 2M because these crystals were the same color as a benzene
solution of 2M, and numerous compounds bearing pseudo C3-sym-
metric molecular structures have been reported to give amorphous
solids.11 However, the infrared (IR) spectrum of IIa was similar to
that of IIb (Fig. S1), indicating that the main component of IIa
was 2D. A comparison of the IR spectra of IIa and IIb revealed sev-
eral differences. As shown in Fig. 3 (top), the IR spectra of IIa and
IIb contained peaks around 1600 cmꢀ1, which nearly merged in
IIa but were well separated in IIb. Notably, the IR spectrum of IIa
contained peaks at 1564 and 1556 cmꢀ1, which did not appear in
IIb. Given that the color of IIa was indicative of the presence of
The molar ratio of 2M to 2D in IIa was estimated based on elec-
tron paramagnetic resonance (EPR) measurements. Before measur-
ing the EPR spectra of IIa, we confirmed that the EPR signal
intensity (integration values) was proportional to the spin number
irrespective of the molecular structure using several standard rad-
ical compounds. The calibration line obtained in this way was used
to evaluate the EPR signal intensity of the weighted amount of IIa,
and the content of 2M in IIa was estimated to be about 9% (See
Electronic Supplementary Information and Fig. S2). The observed
density of IIa (1.029–1.033 g cmꢀ3) was used to determine the
[2D]/[2M]2 value of IIa, which was found to be 8.8 ꢁ 10ꢀ3 (Mꢀ1).
Notably, the molar ratio of 2M to 2D in IIa was different to the value
found in a benzene solution. The dimerization constant K (= [2D]/
[2M]2) in a dilute benzene solution of 2M and 2D was determined
to be 4.5 ꢁ 10ꢀ5 (Mꢀ1),12 which differed considerably from the
[2D]/[2M]2 value in IIa. This result therefore indicated that IIa could
not be regarded as a static structural representation of the 2M/2D
equilibrium in a saturated benzene solution. At this stage we began
to wonder why a mixture of 2M and 2D readily formed the true bin-
ary amorphous solid IIa, whereas mixtures of 1M and 1D, and sev-
eral other related compounds did not show similar properties. This
difference in the behaviors of these materials could be attributed in
part to the enhanced solubility of 2M and 2D in benzene and the
other non-polar solvents mentioned above because of their tert-
butyl substituents. These groups would lead to the rapid solidifica-
tion of solutions of 2M and 2D under highly concentrated condi-
tions. Notably, these conditions would provide
a kinetic
advantageous for the preparation of amorphous solids. This
hypothesis was supported by the fact that hexane and acetone
solutions of 2M and 2D, which were both less soluble in these sol-
vents than they were in benzene, produced IIb.
Another reason for the facile formation of IIa is that these amor-
phous solids are thermodynamically stable. The chemistry of

N. Hayashi et al. / Tetrahedron Letters 58 (2017) 2547–2550
2549
Fig. 2. X-ray structures of Ib (a) and IIb (b). Two of the tert-butyl groups of 2D were disordered in IIb. Oxygen atoms are designated.
solids, they could be behaving in a similar manner to metal
glasses,19 with the smaller molecules (2M) simply filling the voids
between the larger molecules (2D) in an effective manner.
When IIb was vigorously ground in a pestle and mortar, it
immediately became red in color and ultimately formed a deep
purple solid after 90 min.20 The resulting solid IIa0 gave a similar
IR spectrum and XRD pattern to those of IIa, although the XRD pat-
tern did still show the Bragg diffraction peaks of IIb (which failed
to disappear even after several hours of grinding). Moreover, the
concentration of an acetone solution of IIa0 led to the quantitative
formation of IIb. Accordingly, IIa0 should mainly consist of IIa with
a small amount of IIb. Unfortunately, it was not possible to deter-
mine the yield of IIa because of the presence of unreacted IIb.
These results may therefore indicate that the thermodynamic sta-
bilities of IIa and IIb were similar.21 As reported previously, Ib
became red in color when it was ground in a pestle and mortar.8
The resulting solid Ia0 was subjected to IR, ESR, and XRD analyses,
which revealed that it consisted of a mixture of unreacted Ib and
amorphous Ia. Furthermore, the amorphous Ia should consist of
1D with a small amount of 1M, as seen in IIa. The generation of Ia
therefore indicated that this material may be as thermodynami-
cally stable as Ib in a similar manner to IIa versus IIb. However,
the concentration of mixed solutions of 1M and 1D failed to afford
any Ia, most likely because of the lower solubility of 1M and 1D,
which would prevent the rapid solidification of these materials
to give an amorphous solid. The generation of IIa from IIb by grind-
ing could be initiated by the dissociation of 2D to give 2M under
mechanical agitation. In fact, similar radical dissociations have
been reported for the grinding of dimeric molecules in the solid
state, although the products of these reactions are generally crys-
talline in nature.22 Consequently, the molecules would be arranged
without any long-range order to give amorphous solids. The
molecular arrangement would therefore not only involve loca-
Fig. 3. (top) IR spectra (zoom) of IIa (black curve) and IIb (gray curve). (bottom)
Simulated IR absorption peaks of 2 M (black bars) and 2D (gray bars) obtained by
(U)B3LYP/6–31 G(d,p) calculations, where mC@O and mCAOÅ appear at 1670 and
1550 cm–1, respectively.
crystalline materials clearly shows that closely packed structures
are energetically favored over loosely packed lattices, as indicated
by the principle of close packing.13 This principle can be extended
to non-crystalline solids because the packing efficiency of amor-
phous solids, which are thermodynamically less stable than crys-
talline materials, is lower than that of crystals of the same
composition in almost all cases. Moreover, a comparison of binary
systems consisting of the same components but with different
component ratios revealed that the more closely packed amor-
phous solids were the most stable materials.14 Although no general
tendency has been observed in the packing efficiency of amor-
phous solids compared with their corresponding crystal forms,
their amorphous areas are known to be less densely packed by
up to 10% compared with the corresponding crystalline areas in
polymers.15 If this was applicable to molecular amorphous materi-
als, then the packing efficiency of IIa (0.711–0.713) would have to
be considered as unexpectedly high given that it is only a little bit
smaller than that of IIb (0.724–0.725). Considering the principle of
close packing, this result therefore indicates that amorphous IIa
was being stabilized in a thermodynamic manner.16 The facile for-
mation of IIa was reminiscent of amorphous alloys17 and metal
glasses,18 which are also densely packed amorphous solids consist-
ing of two or three atoms of different sizes. Although it still
remains unclear why mixtures of 2M and 2D form closely packed
tional exchange (i.e., M(1) ꢀ M(2) + M(3) ? M(3)ꢂ + M(1) ꢀ M(2)
,
where Mꢂand M–M represent the monomer and dimer molecules,
respectively), but would also involve chemical exchange (i.e.,
M(1) ꢀ M(2) + M(3) ? M(1) + M(2) ꢀ M(3)). The occurrence of the lat-
ter of these two processes was strongly supported by the mass
spectrum of a ground mixture of Ia0 and IIa, which not only showed
molecular ion peaks corresponding to 1D and 2D but also contained
peaks corresponding to the cross dimers 3D and/or 4D (Scheme 2).23
The occurrence of a chemical exchange process between the two
components would disrupt the elimination of the other component
(2M) and the appropriate rearrangement of the 2D molecules dur-
ing the normal crystallization process. This could explain why
the grinding of IIb and Ib afforded poorly ordered solids (i.e., amor-
phous solids).

2550
N. Hayashi et al. / Tetrahedron Letters 58 (2017) 2547–2550
References
1. a) Desiraju GR. Angew Chem Int Ed Engl. 1995;34:2311;
b) Vishweshwar P, McMahon JA, Bis JA, Zaworotko MJ.
2006;95:499;
J Pharm Sci.
c) He G, Chow PS, Tan RBH. Cryst Growth Des. 2009;9:4529.
2. Zallen R. The Physics of Amorphous Solids. Weinheim: Wiley-VCH; 1998.
3. a) Newman A, ed. Pharmaceutical Amorphous
Solid
Dispersions. Weinheim: Wiley-VCH; 2015;
b) Hancock BC, Parks M. Pharm Res. 2000;17:397;
c) Babu NJ, Nangia A. Cryst Growth Des. 2011;11:2662.
4. Shirota Y. J Mater Chem. 2005;15:75.
5. In this study, we have used Arabic figures when referring to molecules with M
or D to denote a monomer or dimer, respectively (e.g., 1M, 2M), whereas solids
(amorphous and crystals) have been denoted with Roman figures (e.g., Ib, IIa).
6. Altwicker ER. Chem Rev. 1967;67:475.
7. Manner VW, Markle TF, Freudenthal JH, Roth JP, Mayer JM. Chem Commun.
2008;256.
8. Dimroth K, Berndt A, Perst H, Reichardt C. Org. Synth. Coll.. 1973;5:1130.
9. Dimroth K, Tüncher W, Kaletsch H. Chem Ber. 1978;111:264.
10. There are several definitions for the term glass. Historically and most generally,
the term glass is used synonymously with amorphous. More recently, the term
glass has been used to refer to X-ray amorphous materials that exhibit a glass
transition. From a thermodynamic perspective, the glass state is defined as
being at non-equilibrium, irrespective of the structure.
11. (a) Whitaker CM, McMahon RJ. J Phys Chem. 1996;100:1081;
(b) Bonvallet PA, Breitkreuz CJ, Kim YS, et al. J Org Chem. 2007;72:10051;
(c) Castellanos S, Valesco D, López-Calahorra F, Brillas E, Julia L. J Org Chem.
2008;73:3759;
Scheme 2. Generation of cross dimer 3D and/or 4D by mixed grinding of Ia0 and IIa.
In conclusion, this paper describes the formation of the true
binary amorphous solids IIa consisting of triarylphenoxyl radicals
(2M) and the corresponding dimer (2D). In contrast to other 2,4,6-
triarylphenoxyl systems, 2M and 2D readily formed the amorphous
solid IIa when they were condensed from benzene solutions con-
sisting of 2D with ca. 9% of 2M. The facile formation of IIa could
be attributed to several factors, including the good solubility of
2M and 2D in benzene (resulting from the tert-butyl substituents),
which led to the rapid solidification of the substrate compared
with several other solvents (e.g., hexane), and the closely packed
structure of IIa. Chemical exchange between 2M and 2D could also
be involved in this process, playing an important role in the con-
version of IIb to IIa by solid-state grinding. Further research
towards understanding the solidification behavior of related com-
pounds is currently underway in our laboratory, together with
research aimed at elucidating the mechanisms responsible for
the formation of amorphous solids.
(d) Castellanos S, Gaidelis V, Jankauskas V, et al. Chem Commun. 2010;46:5130.
12. Dimroth K, Kalk F, Sell R, Schlömer K. Justus Liebigs Ann Chem. 1959;624:51.
http://dx.doi.org/10.1002/jlac.19596240107.
13. Kitaigorodoskii AI. Acta Cryst. 1965;18:585.
14. Li Y, Guo Q, Kalb JA, Thompson CV. Science. 2008;322:1816.
15. Dlubek G, Stejny J, Lüpke Th, et al. J Polym Sci Part B Polym Phys. 2001;40:65.
16. Schwarz RB, Johnson WL. Phys Rev Lett. 1983;51:415.
17. Inoue A, Ohtera K, Tsai AP, Masumoto T. Jpn J Appl Phys. 1988;27:L280.
18. a) Wang WH, Dong C, Shek CH. Mater Sci Eng. 2004;R44:45;
b) McHenry ME, Willard MA, Laughlin DE. Prog Mater Sci. 1999;44:291.
19. Miracle DB, Sanders WS. Philos Mag. 2003;83:2409.
20. Generation of IIa from IIb occurred only when IIb was ground vigorously. In
contrast, when IIb was ground gently, no change except pulverization occurred.
21. However, DSC analysis (heating rate = 5 °C minꢀ1) revealed that there was no
thermal phase transition from IIb to IIa (vice versa), whereas IIb and IIa
decomposed at Tonset = 167 and 128 °C, respectively.
22. Kawano M, Sano T, Abe J, Ohashi Y. Chem Lett. 2000;1372.
23. Formatino of 3D and/or 4D should not be attributed to thermal reaction under
the mass spectrometry conditions used. In the mass spectrometry the sample
was kept at lower than 80 °C during the experiment, while DSC analysis
(heating rate = 5 °C minꢀ1) of the mixture of Ia0 and IIa showed neither endo-
nor exotherm at that temperature.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2017.05.
040.