Electron Spin Resonance Study of Molecular Orientation and

Subscriber access provided by Kaohsiung Medical University

A: Kinetics, Dynamics, Photochemistry, and Excited States

ESR Study of Molecular Orientation and Dynamics of Phenyl Imino and Nitronyl Nitroxide Radicals in Organic 1D Nanochannels of Tris(o-phenylenedioxy)cyclotriphosphazene Hirokazu Kobayashi, Takanori Mori, Yuka Morinaga, Etsuko Fujimori, Kento Akiniwa, and Fumiyasu Iwahori J. Phys. Chem. A, Just Accepted Manuscript • Publication Date (Web): 07 Jun 2018 Downloaded from http://pubs.acs.org on June 7, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

ESR Study of Molecular Orientation and Dynamics of Phenyl Imino and Nitronyl Nitroxide Radicals in Organic 1D Nanochannels of Tris(ophenylenedioxy)cyclotriphosphazene Hirokazu Kobayashi1*, Takanori Mori2, Yuka Morinaga2, Etsuko Fujimori2, Kento Akiniwa3, Fumiyasu Iwahori2 1

Faculty of Arts and Sciences at Fujiyoshida, Showa University, 4562, Kami-yoshida, Fujiyoshida-shi, Yamanashi, 403-0005, Japan. 2

Department of Chemistry, College of Humanities and Sciences, Nihon University, 3-25-40, Sakura-jo-sui, Setagaya-ku, Tokyo, 156-8550, Japan. 3

Graduate School of Integrated Basic Sciences, College of Humanities and Sciences, Nihon University, 3-25-40, Sakura-jo-sui, Setagaya-ku, Tokyo, 156-8550, Japan.

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 48

ABSTRACT Imino

and nitronyl

nitroxide (IN and NN, respectively) radicals such as

phenyliminonitroxide (PhIN) and phenylnitronylnitroxide (PhNN), respectively, were dispersed in the organic 1D nanochannels of tris(o-phenylenedioxy)cyclotriphosphazene (TPP). Electron spin resonance (ESR) measurements were conducted on these inclusion compounds (ICs) in the temperature range of 4.2-300 K. The modulated-septet ESR spectra of TPP ICs using PhIN observed in the range of 165-258 K were reproduced with the EasySpin program package using a model in which some of the PhIN molecules underwent uniaxial rotational diffusion in the TPP nanochannels around the molecular long axis corresponding to the principal y-axis of the gtensor. On the other hand, for the TPP IC using PhNN, complicated ESR spectra were observed, which were not consistent with the modulated quintet observed in solution or in the fast-motion limit in solids. These spectra were reproduced by the superposition of a quintet originating from rotational diffusion of PhNN molecules and a septet based on rotational diffusion of PhIN molecules generated in the synthetic process. The rotational diffusion activation energies of PhIN and PhNN in the TPP nanochannels were estimated to be 19 and 45 kJ mol-1 using Arrhenius plot, respectively. These were consistent with that for NN radicals in the 1D nanochannels of 2,4,6-tris(4-chlorophenoxy)-1,3,5-triazine (CLPOT) (37-54 kJ mol-1) with a larger pore diameter than TPP reported in our previous study, or with that for 4-substituted2,2,6,6-tetramethyl-1-piperidinyloxyl (4-X-TEMPO) in TPP nanochannels (5-26 kJ mol-1) with regard to molecular size, or host-guest or guest-guest interactions. These results indicate that not only the NN group but also IN may be used for the clarification of chemical or biological structures of nanomaterials such as nano-sized cavities.


2

Page 3 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Recently, a wealth of studies on self-assembled materials having 0-3 dimensional nanospace have been reported.1-7 One-dimensional (1D) porous materials with regularly shaped periodic cavities such as 1D nanochannels are frequently used as nanosized molecular templates. Using 1D nanochannels as molecular templates, many of anisotropic functional inclusion compound have been constructed.8-10 The 1D alignment of paramagnetic molecules using 1D nanochannels as templates is expected to give rise to anisotropic magnetic properties.11-15 Such materials have attracted attention in the development of new molecular magnets,11 molecularsized devices,16 medicines with magnetic markers,17 and for applications in magnetic resonance spectroscopy.18 The formation of 1D molecular chains of 2,2,6,6-tetramethyl-1-piperidinyloxyl (TEMPO) radicals was achieved using 1D nanochannels of tris(o-phenylenedioxy)cyclotriphosphazene crystals (TPP19,20; Scheme 1a) ([(TPP)2-(TEMPO)1.0]).21,22 In electron spin resonance (ESR) measurements, an isotropic line profile showing the 1D spin diffusion behavior was observed above 139 K.

In addition, the molecular motion of TEMPO radicals dispersed in TPP

nanochannels by similar-sized non-radical spacers was found to undergo anisotropic rotational diffusion around the principal y axis of the g-tensors of TEMPO molecules perpendicular to the molecular long axis above 108 K.23 These results suggested the possibility of correlating the magnitude and/or dimensionality of inter-spin exchange/dipolar interactions of [(TPP)2(TEMPO)1.0] with the correlation time and/or activation energy of molecular motion of TEMPO radicals in TPP nanochannels.


3


Page 4 of 48

Scheme 1. Chemical structures of host and guest compounds used in this study.

This suggests the possibility of developing [(organic 1D porous material)-(organic radical)] inclusion compounds (O1DP-ORICs) for probing the magnetic exchange between radicals by the appropriate choice of host and guest materials for the control of molecular orientation and dynamics of the guest radicals in the nanochannels. In fact, such ICs were reported, in which 3D exchange interaction was observed by the use of a porous material with larger pore diameter than TPP.24 The inclusion of several organic radicals in TPP nanochannels


4

Page 5 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and the molecular orientation and dynamics of guest molecules have been confirmed using ESR spectroscopy25-29 and spectral reproduction by means of computer simulation.30-32 In addition, the molecular orientation and dynamics of organic radicals in confined nanospace such as not only in 1D nanochannels but also in 0D cavities in cyclodextrins33 or in micelles.34 1D nanochannels of 2,4,6-tris(4-chlorophenoxy)-1,3,5-triazine (CLPOT)35 crystals have also been used for the confinement of several 4-substituted-TEMPO derivatives (4-XTEMPO)36,37 and 4-substituted-phenylnitronylnitroxide (4-XPNN) radicals.38 Especially when 4-XPNN were used as guest radicals, the possibility of developing a new ESR spin probe technique using NN groups were suggested by using the molecular orientation and dynamics of NN radicals in confined spaces determined by ESR measurements and spectral simulation. In the synthesis of O1DP-ORICs, CLPOT crystals are preferable to TPP as a host material to include many kinds of organic radicals due to: (1) its pore diameter, which is comparable to the molecular cross section of many stable nitroxide or NN radicals (ca. 1.1-1.3 nm in CLPOT, but 0.45-0.9 nm in TPP), and (2) its favorable polarity (radicals frequently have polar substituent groups). However, if TPP can also include various organic radicals, the variety of designs of O1DP-ORICs could be substantially extended. In this study, a small inclusion amount of PhNN (Scheme 1b) in TPP nanochannels was achieved by means of co-inclusion with spacers according to the synthetic process reported in reference 38.

In addition, a small inclusion amount of 4-phenyliminonitroxide (PhIN39-45;

Scheme 1c) radical was also attained in the TPP nanochannels by a similar synthetic process. Temperature-dependent ESR measurements were used to determine the g and hyperfine (A) tensors for PhNN and PhIN in TPP nanochannels, and their molecular orientation and dynamics were clarified in detail by spectral reproduction using EasySpin.31,32 In particular, the g and A


5


Page 6 of 48

tensors for PhIN have rarely been reported,45 let alone the molecular dynamics in confined spaces. This paper describes these remarkable results.

Experimental 2.1 Chemicals TPP (solid, colorless) was synthesized according to a previously reported method.19 PhNN (solid, indigo blue) and N-phenyl maleimide (N-PhMI; Scheme 1d) (thin plate-like crystal, pale green) were purchased from Tokyo Chemical Industry Co. Ltd.

PhIN (orange oily

compound around room temperature; low boiling point40) was synthesized according to a previously reported method.39 All chemicals were used without further purification. 2.2 Sample preparation Samples were prepared according to the following process. The inclusion of PhIN or NN in the TPP nanochannels was attained using N-PhMI as a spacer since it has a similar molecular shape to the IN and NN radicals. [(TPP)2-(PhIN)x/(N-PhMI)y] IC (1) was prepared using the following procedure. 1.2×10-3 mol of N-PhMI and 7.8×10-5 mol of PhIN were mixed in 2 mL of mesitylene, and the solution was poured into 3 mL of 0.15 mol L-1 TPP mesitylene solution at 383 K. A colorless powder was obtained as the product after cooling the solution to room temperature. The possibility of


6

Page 7 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the adsorption of a part of PhIN molecules on the surface of TPP crystals were given (see section 3.1 and 3.2). [(TPP)2-(PhNN)x/(PhIN)y/(N-PhMI)z] IC (2) was prepared using the following procedure. 1.2×10-3 mol of N-PhMI and 8.6×10-5 mol of PhNN were mixed in 2 mL of mesitylene, and the solution was poured into 3 mL of 0.15 mol L-1 TPP mesitylene solution at 383 K. A colorless powder was obtained as the product after cooling the solution to room temperature.

The

possibility of the generation of PhIN from a part of PhNN by reduction40 were given (see section 3.1 and 3.3). [(TPP)2-(N-PhMI)y] IC (3) was prepared using the following procedure. 1.2×10-3 mol of NPhMI was mixed into 3 mL of 0.15 mol L-1 TPP mesitylene solution at 383 K. A pale yellow powder was obtained as the product after cooling the solution to room temperature. After heating for an hour at 403 K, the sample color was changed into colorless. Characterization of all compounds was conducted using powder X-ray diffraction (XRD), ESR spin concentration measurements, elemental analysis (EA) and thermogravimetric analysis differential thermal analysis (TG-DTA). After sample characterization, variable-temperature ESR measurements of the ICs were conducted. Compounds 1-3 were stable for at least a half year in air. The weight loss during TG-DTA was observed only for 2 and 3 due to the decomposition of 1 by heating (see below). The color change of 3 in the synthetic process may be caused by the desorption of N-PhMI on the surface of TPP crystals. The DTA peak of bulk N-PhMI observed at 88 oC reported in reference 38 was not observed in 1-3. These indicate that bulk N-PhMI rarely adsorbed on the surface of the TPP crystals of 1-3 (see below).


7


Page 8 of 48

2.3 Instrumentation Powder XRD analyses of all samples were conducted at room temperature using a diffractometer (Rint 2100, Rigaku Corp.) with graphite monochromated Cu Kα radiation (λ = 0.15418 nm). Data were collected in the θ–2θ scan mode using a 2θ scan rate of 2° min-1 in the 2θ range of 5–50°. TG-DTA measurements were conducted on all samples using a thermogravimetric apparatus (TG-DTA8122, Rigaku Corp.). 10 mg of powder specimens and Al2O3 powder as a reference were placed in Pt sample pans, set in the apparatus, and heated between room temperature and 300 °C at a heating rate of 10 °C/min under a N2/Ar gas atmosphere. ESR spectra were recorded using an X-band spectrometer (JES-FA300, Jeol) with a TE011 cylindrical cavity resonator at temperatures from 4.2 to 300 K. 2–3 mg of powdered specimen was packed in a commercial quartz glass ESR tube (270 mm long, 5 mm o.d.) and then sealed under a He atmosphere at 4 kPa. Thermal equilibrium of the sample was achieved by waiting 5– 10 min after each temperature change.

Signal reproduction was also confirmed for both

increasing and decreasing temperatures. The X-band (e.g., 9.072091 GHz) microwave power was set to 0.001-1 mW under non-saturated conditions. When measurements were performed below 20 K, saturation occurred above about 0.005 mW. The microwave power was set at 0.36 mW for measurements in the range 20-300 K under non-saturated conditions. The magnetic field range and field sweep rate were set at 323±15 mT and 30 mT/8 min, respectively. 100 kHz field modulation with an amplitude of 0.1 mT was used. The spin concentration was measured by comparing the double integral of the cw-ESR spectra of weighed samples with that of a


8

Page 9 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TEMPOL standard solution via the signal intensity of a manganese standard. All measurements were conducted with the same loading of the cavity. The estimated error was 30%. Spectral simulations of compounds 1 and 2 were performed using the Pepper program for solid-state spectra and the Chili program software package for anisotropic slow-motion cw-ESR spectra (EasySpin 5.2.16, ETH Zürich).31,32 Calculations for IN and NN radicals could be completed using a PC with more than 8 GB of RAM. Chili required more memory to run the spectral reproduction of NN radicals; therefore, the “Allocation” function of Opt should be set to “Opt.Allocation = [4e7 2e5]”.32

2.4 Procedure to obtain the rotation axes of the radical in the nanochannels The Hamiltonian of free radicals in 1D nanochannels follows that reported previously.38 NN or IN radicals oriented in the 1D nanochannels are characterized by the spin Hamiltonian: ෡ = ߚ௘ ࡮ ∙ ࢍ ∙ ܵመ + ܵመ ∙ ෍ ࡭࢏ ∙ ‫ܫ‬෡ప + ߚ௡ ݃௡ ∙ ࡮ ∙ ෍ ‫ܫ‬෡ప ‫ܪ‬ ௜ୀଵ,ଶ

௜ୀଵ,ଶ

(1) where βe, B, g, Ai, βn, gn, ܵመ, and ‫ܫ‬෡ప are the Bohr magneton, the laboratory magnetic flux density vector, the electron spin g tensor, the hyperfine tensor for the ith

14

N nucleus in the NN or IN

radical (i = 1 or 2), the nuclear magneton, the nuclear spin g-factor, the electron spin operator, and the nuclear spin operator of the ith 14N nucleus in the NN or IN radical, respectively. In this study, the principal axes of the NN radicals are defined as follows: the z-axis for PhNN is perpendicular to the molecular plane of the NN group; the y-axis is parallel to the bond between the 2-position carbon atom of the NN group and the phenyl ring; and the x-axis is


9


Page 10 of 48

perpendicular to the yz plane.38 Therefore, the g tensor for the NN radical is determined as follows. The unpaired electron of the NN group is associated with a π orbital. The lowest component of the g tensor should be observed perpendicular to the molecular plane of the NN group, i.e., the principal z axis direction, and should be near 2.0023.38,45,46 In molecular orbital calculations, gxx >gyy >gzz for the NN radical (i.e., the intensity of the gxx component is observed at the lowest magnetic field side). With respect to the A tensor for the NN radical, the unpaired electron occupies the π orbital composed of the 2pz orbitals of the nitrogen atoms with considerable spin density, such that Azz is the greatest component,46 whereas Axx and Ayy are expected to be less than Azz. Therefore, the A tensor for NN radicals is quite anisotropic. Although the local axes of the A tensors of the two nitrogen nuclei of the NN radical are expected to be different, it was assumed that the unpaired electron interacts with the two nitrogen nuclei with an averaged A tensor.38,46 This is solely to make the analysis simple and easy. In this approximation, the principal axes of the g and A tensors for the NN radical will be coincident. The principal axes of the g-tensor of the IN radical are assumed to be similar to the NN radical and also consistent with NN due to the higher spin density on N-O.40,45 However, note that A1IN ≠ A2IN for IN radicals ((a1IN) iso > (a2IN)iso), where A1IN is the hyperfine tensor of the nitroxide nitrogen of the IN groups and A2IN is that of the 3-position nitrogen. Since the xx and yy components of A1IN and A2IN are much smaller than the zz component, the principal axes of the g, A1IN, and A2IN tensors of for the IN radical are also assumed to be coincident for simplified calculations according to the procedure for NN radicals. An example of a rigid-limit ESR spectrum for the NN radical has previously been reported.38,46 A typical rigid-limit ESR spectrum for IN radicals simulated using the Pepper


10

Page 11 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


program software package (EasySpin 5.2.13; ETH Zürich31,32) for calculation of solid-state cwESR spectra is depicted in Figure 1a. The seven major hyperfine interaction lines associated with the inequivalent nitrogen atoms with different hyperfine coupling constants (in many cases, A1zzIN ~ 2A2zzIN) are shown on the horizontal axis in Fig. 1a for convenience. The peak heights of these supplementary lines are proportional to their peak intensities. Although lines 1, 2, 5, 6, and 7 are well-resolved, the spectrum between lines 3 and 4 is more complicated. These are caused by gxx and gyy components that are larger than gzz, and by A1xxIN, A1yyIN, A2xxIN and A2yyIN components that are much smaller than A1zzIN or A2zzIN.


11


Page 12 of 48

Figure 1. Simulated ESR spectra of (a) typical powder rigid IN radicals using the Pepper program in EasySpin31,32 and (b) IN radicals modulated on the basis of anisotropic rotational diffusion motion around the principal z (top), x (middle) and y (bottom) axes of the g-tensor using the Chili program within the slow-motion regime with respect to the ESR time scale. Bars on the horizontal axis in (a) indicate the seven major hyperfine coupling interactions by inequivalent nitrogen atoms (A2zzIN = 1.2 mT and A1zzIN = 2.4 mT; see Table 1). The g, A1IN and A2IN tensors for the IN radical were experimentally estimated as follows. gzz and A2zz were respectively set as default from the seven major hyperfine coupling lines of the measured ESR spectrum for the isolated IN radical at low temperature, in which the molecular


12

Page 13 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


motion of the IN radical is thought to be frozen. Next, 2A1zzIN was set as default from peaks 2 and 6 corresponding to lines 2 and 6 at the bottom of Fig. 1a. gxx and gyy, and Axx and Ayy are difficult to determine directly from the ESR spectra; therefore they were first estimated visually from the spectral reproduction of the experimental ESR spectra using EasySpin. The final values for each component of the g, A1IN and A2IN tensors of the IN radical was estimated by step-bystep variation of these components or gii, (A1IN)ii and (A2IN)ii (i = x, y, or z), line width parameters, and signal amplitude, so as to minimize the sum of the squares of the difference between the experimental and simulated spectra. In EasySpin, the line shape is expressed by a Voigt function, which is defined by Gaussian broadening (convolution) of a Lorentzian line.38,47 In the present analysis, the full width at half maximum (FWHM) was taken as the line width parameter. In Fig. 1a, the Gaussian and Lorentzian line width components were taken as 0.10 and 0.40 mT, respectively. The g, A1IN and A2IN tensors used for the reproduction in Fig. 1 are based on the results shown in Table 1 as well as those of PhNN radicals in some cases.38,45,46 For reproduction of ESR spectra at higher temperature in which molecular motion of the radicals is excited in the 1D nanochannels, it was assumed that the rotation axis of PhNN or IN radicals is defined by polar and azimuthal angles, θ and φ, respectively, in the principal axis system of the g-tensor for NN or IN radicals. In the domains of 0 ≤ θ ≤ π and 0 ≤ φ ≤ 2π, only the (θ0, φ0) values are given although four possible rotation axes were derived.38


13


Page 14 of 48

Table 1. Tensor components of g, A1IN and A2IN for PhIN and g and A for PhNN under various conditions. A1xx/A2xx / A1yy/A2yy mT / mT

A1zz/A2zz / mT

State of radicals

gxx

gyy

gzz

PhIN glassy toluene *1

2.00932

2.00604

2.00222

PhIN in TPP (this study; 1)

2.0103

2.0050

2.0028

0.02/0.01

0.01/0.02 2.36/1.16

State of radicals

gxx

gyy

gzz

Axx / mT

Ayy / mT

Azz / mT

PhNN in rigid Duco Cement *2

2.0127

2.0068

2.0028

0.52

0.52

1.80

PhNN in glassy toluene *1

2.0100

2.00653

2.00210

----

----

1.86

PhNN in CLPOT*3

2.0118

2.0075

2.0031

0.34

0.12

1.74

PhNN in TPP (this study; 2)

2.0107

2.0086

2.0031

0.01

0.01

2.13

*1

Ref. 45.

*2

Ref. 46.

*3

----

----

2.32/1.28

Ref. 38.

The anisotropic molecular rotational diffusion in the organic 1D nanochannels was dealt according to the previous study of NN radicals in CLPOT nanochannels.38 The values of θ and φ of NN and IN radicals were estimated by varying τR, line width parameters and signal amplitude so as to obtain the best fit for the ESR spectrum at the highest temperature, at which the molecular motion of guest radicals in the 1D nanochannels is in the slow-motion regime (10-6 s > τR > 10-9 s) on the ESR time scale. By using these θ and φ values, τR at each temperature was estimated similarly by varying τR, the line width parameters, and the signal amplitude.

3. RESULTS AND DISCUSSION


14

Page 15 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3.1 Powder XRD patterns and guest inclusion amounts of TPP ICs including PhIN and PhNN The crystal structure and the inclusion amount of guest compounds of 1-3 were examined according to the previous study of NN radicals in CLPOT nanochannels.38 Figure 2 shows powder XRD patterns for (a) guest-free TPP, (b) 1 ([(TPP)2-(PhIN)0.002/(N-PhMI)1.0]), (c) 2 ([(TPP)2-(PhNN)x/(PhNN)y/(N-PhMI)1.0]) (x ~ 5×10-4, y ~ 2×10-4), (d) 3 [(TPP)2-(N-PhMI)1.0], and (e) bulk N-PhMI at room temperature. (see below regarding the compositions, and note that sample names in Fig. 2 are shown in abbreviated form). The vertical bars under Figs. 2b-d are estimations assuming that 1-3 belong to the same space group as guest free TPP (a, P63/m)19,20 and that they have different cell parameters (Table 2). Since the results in Figs. 2b-d are consistent with these assumptions, TPP nanochannels are expected to exist in 1-3.

As a

reference, the estimated cell parameters for crystals of a few TPP ICs using nitroxide radicals are also listed in Table 2. Since peaks for bulk N-PhMI are not observed in 1-3 shown in in Fig. 2bd, bulk N-PhMI was determined to be absent, at least on the surface of TPP crystals of 1-3. These results were also supported by TG-DTA (see section 2.2). Although no peaks for bulk PhIN in 1 or bulk PhNN in 2 are not observed in Fig. 2b or c, note that PhIN is not necessarily a solid at room temperature (see section 2.1 and below).


15


Page 16 of 48

Figure 2. Powder XRD patterns of (a) guest-free TPP, (b) 1 ([(TPP)2-(PhIN)0.002/(NPhMI)1.0]), (c) 2 ([(TPP)2-(PhNN)x/(PhIN)y/(N-PhMI)1.0]) (x ~ 5×10-4, y ~ 2×10-4), (d) 3 [(TPP)2-(N-PhMI)1.0], and (e) bulk N-PhMI at room temperature. The vertical bars under the b-d patterns are estimates assuming that b-d belong to the same space group as guest-free TPP (a, P63/m) and that they have different cell parameters. Note: sample names are shown in abbreviated form.


16

Page 17 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 2. Cell parameters for crystals of a few TPP ICs using nitroxide radicals and 1-3. Compounds

a / nm

c / nm

guest free TPP*1

1.1454

1.0160

[(TPP)2-(TEMPO)0.02/(TEMP)0.98]*2 1.220

0.995

[(TPP)2-(DTBN)x/(pivalone)1.1]*2 1.198

0.998

1

1.185

1.002

2

1.183

1.002

3

1.186

1.003

(x = 4 × 10−4)

*1

Ref. 20.

*2

Ref. 27.

Experimental results for EA, ESR spin concentration, and desorption amounts from TG measurements are given in Table 3. The bold italic entries under the Radicals and Spacers headings show the estimated fraction of radical and/or spacer molecules in the TPP unit cell formed by 2 TPP molecules, which is consistent with the experimental results. The theoretical values of the estimated fraction in italics are listed under each experimental result. According to the ESR results for 1, some PhIN may be adsorbed on the surface of the TPP crystals (see below). Therefore, the inclusion amount of PhIN in the TPP nanochannels is expected to be less than 0.006 in the TPP unit cell. During TG-DTA of 1, the weight of the sample suddenly decreased around 204 oC, and the anomaly of DTA was observed.

These indicate the

decomposition of the guest radicals by heating before the desorption of guest compounds (see section 2.2). In addition, the ESR results for 2 suggest that not only PhNN but also PhIN radicals are included in the TPP nanochannels in the synthetic process (see section 2.2 and 3.3). Therefore, the inclusion amount of guest radicals in 2, 7×10-4, is the summation of PhNN and PhIN. The TG measurements for 2 and 3 support the other results obtained for them (needless to


17


Page 18 of 48

say, no ESR signal was obtained in 3). In summary, compounds 1-3 are respectively assigned as follows: (1) [(TPP)2-(PhIN)0.002/(N-PhMI)1.0] (see section 3.2), (2) [(TPP)2-(PhNN)x/(PhIN)y/(NPhMI)1.0] (x ~ 5×10-4, y ~ 2×10-4, see section 3.3), and (3) [(CLPOT)2-(N-PhMI)1.0].

Table 3. Experimental results for EA, ESR spin concentration, and desorption amount from TG measurements. Bold italic entries under Radicals and Spacers show the estimated fraction of radical and/or spacer molecules in the TPP unit cell formed by 2 TPP molecules, which is consistent with the experimental results. The theoretical values of the estimated fraction in italics are listed under each experimental result. Compounds

1

ESR spin The desorption concentration amount in TG measurements/% / g-1

Radicals

Spacers

H/%

C/%

N/%

PhIN

N-PhMI

3.00

50.46

8.72

3×1019

decomposition

0.006

1.0

2.87

50.64

8.98

3×1019

15.9

N-PhMI

2.83

50.70

8.90

4×1017

14.2

1.0

2.86

50.62

8.98

4×1017

15.9

N-PhMI

3.05

50.46

8.79

----

13.3

1.0

2.86

50.62

8.98

----

15.9

PhNN 2 and PhIN 7×10-4 3

EA

----

3.2 ESR spectra for PhIN radicals in the TPP nanochannels Figure 3 shows temperature-dependent ESR spectra for 1 ([(TPP)2-(PhIN)0.002/(N-PhMI)1.0]). All spectra were normalized to have the same maximum peak-to-peak height. Each green, yellow and red line indicates a rotational diffusion and rigid-limit component of PhIN in 1 reproduced using EasySpin,31,32 along with their superposition. The spin concentration for each sample remained unchanged after the temperature-dependent ESR measurements. The spectra taken at less than 165 K were well reproduced based on a rigid-limit powder pattern for isolated


18

Page 19 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


PhIN. The spectrum for 1 at 56 K shown in Fig. 3 was reproduced using the g and A tensor components shown in Table 1, and the respective Gaussian and Lorentzian line width components were 0.31 and 0.30 mT. The values of the g and A tensors for 1 estimated using EasySpin were consistent with the previous results shown in Table 1. Therefore, in 1, the PhIN radicals are expected to be adequately isolated in the TPP nanochannels by spacers and/or on the surface of the TPP crystals and the molecular motion is frozen on the ESR time scale (i.e., τR > 10-6 s).


19


Page 20 of 48

Figure 3 Temperature-dependent ESR spectra for 1 ([(TPP)2-(PhIN)0.002/(NPhMI)1.0]). All spectra were normalized to have the same maximum peak-to-peak height Each green, yellow and red line indicates rotational diffusion and rigid-limit components of PhIN in 1 reproduced using the EasySpin program package,31,32 and their superposition. If PhIN radicals are included in the TPP nanochannels, temperature-dependent ESR spectra are expected to be observed according to previous reports.38 However, the spectra above 165 K of 1 were not reproduced by a single anisotropic rotational diffusion components but by a superposition of anisotropic rotational diffusion and rigid-limit components. From 258 K to room temperature, the rotational diffusion and rigid-limit components were temperatureindependent within the experimental error. The anisotropic rotational diffusion component of 1


20

Page 21 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in the temperature range from 165 K to 258 K was reproduced as a septet of two 14N atoms of the IN group of PhIN (I = 1 both together, but with different hyperfine coupling constants) modulated by the molecular motion in the slow region with respect to the ESR time scale. These results therefore suggest that some PhIN radicals may be included in the TPP nanochannels, but that additional PhIN radicals may be on the surface of the TPP crystals. In diluted solutions of PhIN or in nanospace in the fast-motion limit on the ESR time scale, seven major hyperfine splitting lines associated with the inequivalent nitrogen atoms of IN with A1zzIN ~ 2A2zzIN were observed with an intensity ratio of 1:1:2:1:2:1:1, whereas in the molecular motion in the slow region of the ESR time scale, the ESR spectra of IN radicals were modulated as shown in Fig. 1b (see section 2.4 and reference 45) depending on the orientation of the rotation axis. The ESR spectrum of 1 at 258 K was seemingly similar to the reproduction by the anisotropic rotational diffusion around the principal y-axis of the g-tensor shown in the bottom of Fig. 1b. However, upon closer inspection, the intensity of the center peak (the 4th from the left side) at 258 K in Fig. 3 was nearly equal to the right side peak (5th from the left side), and that of the 3rd peak from the left side was slightly larger than that of the 4th or 5th peak from the left side. When a rigid-limit component attained at low temperature was superposed, the spectral reproduction was better, as shown Fig. 3. In fact, the modulated-septet of the ESR spectrum of 1 at 258 K was approximately reproducible assuming the anisotropic rotational diffusion around the axis was tilted by (θ, φ) = (90°, 73°) to the principal axis system of the g-tensor and parallel to the principal y-axis, with Gaussian and Lorentzian line width components and τR of 0.17 and 0.036 mT, and 1.3×10-8 s, respectively. These results show that the rotation axis was also approximately parallel to the molecular long axis of the PhIN molecules (i.e., the molecular long axis is parallel to the channel axis of the TPP nanochannels, if it is assumed that the rotation axis


21


Page 22 of 48

is parallel to the channel axis). Every ESR spectrum collected in the range from 165 to 239 K was also reproduced by a set of methods. The ratio of the anisotropic rotational diffusion component to the whole integrated intensity was about 22% in the temperature range from 165 K to room temperature. Therefore, the inclusion amount of PhIN in 1 was determined to be x ~ 0.002; [(TPP)2-(PhIN)0.002/(N-PhMI)1.0] (see also section 3.1).

According to the spectral

reproduction process, at least some PhIN is expected to be included in the TPP nanochannels although other PhIN radicals in 1 may be adsorbed on the surface of the TPP crystals. In the XRD and TG-DTA results for 1, no remarkable XRD patterns or DTA peaks before decomposition of bulk PhIN were observed (see also section 2.2 and 3.1). This may indicate that the IN molecules were not crystallized but isolated on the surface of the TPP crystal.

3.3 ESR spectra for PhNN radicals in the TPP nanochannels Figure 4 shows temperature-dependent ESR spectra for 2 ([(TPP)2-(PhNN)x/(PhIN)y/(NPhMI)1.0] (x ~ 5×10-4, y ~ 2×10-4)). All spectra were normalized to have the same maximum peak-to-peak height. The spin concentration of each sample remained unchanged after the temperature-dependent ESR measurements. In the temperature range below 131 K or above 253 K, the ESR spectra were temperature-independent within the experimental error.


22

Page 23 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4 Temperature-dependent ESR spectra for 2 ([(TPP)2-(PhNN)x/(PhIN)y/(N-PhMI)1.0] (x ~ 5×10-4, y ~ 2×10-4). All spectra were normalized to have the same maximum peak-topeak height. Each green, yellow and red line indicates rotational diffusion components of PhNN and PhIN in 2 reproduced using EasySpin,31,32 and their superposition. The asterisks in the spectrum at 253 K indicate the peaks corresponding to PhNN, whereas the diamonds indicate the peaks corresponding to PhIN.

We attempted to reproduce the spectra of 2 using a single rigid-limit or rotational diffusion component of PhNN, or a superposition of them according to the reproduction process of 1 (see above) over the entire temperature range. In diluted solutions or in nanospaces in the fast-motion


23


Page 24 of 48

limit of the ESR time scale, the ESR spectra of PhNN show a 1:2:3:2:1 quintet with aiso ~ 0.7 mT.38 Therefore, around room temperature, the ESR spectra of 2 are expected to be comparable to those for the fast-motion limit. However, the spectra of 2 shown in Fig. 4 are consistent with none of these and are more complicated over the entire temperature range, even above 253 K at which the spectra are in the fast-motion limit on the ESR time scale. Upon closer inspection, since the peaks indicated by asterisks seemingly represent 0.7 mT of hyperfine splitting with an intensity ratio similar to the fast-motion region, these are expected to originate from PhNN radicals. In addition, the peaks marked by diamonds in the spectrum at 253 K represent 0.4 mT of hyperfine splitting. Here, if A1zzNN ~ A1zzIN ~ 2A2zzIN (see section 2.4), these peaks are expected to originate from PhIN. Figure 5 shows ESR spectra reproduced by a superposition of a quintet of NN groups modulated by rotational diffusion around the principal y axis of the g-tensor and the modulated septet of IN groups shown in Fig. 3 at a reasonable fraction. All spectra in Fig. 5 were normalized to have the same maximum peak-to-peak height. The Gaussian and Lorentzian line width components, and the rotational diffusion correlation time, τR, for the NN and IN spectra used in Fig. 5 were respectively 0.1 and 0.2 mT, and 6×10-10 s in the spectra of NN groups (green) and 0.2 and 0.1 mT, and 1×10-8 s in those of IN groups. In Fig. 5, the spectrum with the intensity ratio of NN:IN = 60:40 was more comparable to the experimental spectrum at 253 K in Fig. 4 than other cases with different component ratios. Therefore, the spectra of 2 over the entire temperature range shown in Fig. 4 were expected to be reproducible by a superposition of rotational diffusion components of PhNN and PhIN at NN:IN ~ 3:2 or 2:1. In Fig. 4, each green, yellow and red line indicates a corresponding PhNN or PhIN component in 2 reproduced using EasySpin,31,32 and their superposition.

The spectrum for 2 at 52 K shown in Fig. 4 was


24

Page 25 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reproduced using the g and A tensor components for PhNN and PhIN shown in Table 1, and the respective Gaussian and Lorentzian line width components were 0.68 and 0.76 mT for PhNN, and 0.048 and 1.50 mT for PhIN. The values of the g and A tensors for PhNN and PhIN in 2 estimated using EasySpin were consistent with the previous results shown in Table 1 and the previous section (i.e., the aiso of PhNN or PhIN in Table 1 was also comparable to the visual estimation). Therefore, PhNN and PhIN radicals in 2 are expected to be adequately isolated in the TPP nanochannels by the spacers, and the molecular motion is frozen with respect to the ESR time scale (i.e., τR > 10-6 s). Note that the compositions of PhNN and PhIN are described by the summation x+y = 7×10-4 (see section 3.1).


25


Page 26 of 48

Figure 5 The ESR spectra reproduced by the superposition of rotational diffusion components of NN and/or IN groups around the principal y-axis of the g-tensor at a reasonable component rate with EasySpin.31,32 The Gaussian and Lorentzian line width components, and the rotational diffusion correlation time τR were respectively taken as 0.1 and 0.2 mT, and 6×10-10 s in the ESR spectra of NN groups (green) and 0.2 and 0.1 mT, and 1×10-8 s in those of IN groups. The ESR spectrum at 253 K is expected to be reproduced by a superposition of the rotational diffusion components of NN groups around the principal y axis of the g-tensor reported in


26

Page 27 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reference 38 and those of IN groups shown in Fig. 3. The spectrum at 253 K in Fig. 4 was approximately reproducible on the assumption of a superposition of anisotropic rotational diffusion around the axis tilted by (θ, φ) = (90°, 89°) to the principal axis system of the g-tensor for PhNN with Gaussian and Lorentzian line width components and τR of 0.03 and 0.17 mT, and 6.1×10-10 s, and by (θ, φ) = (90°, 71°) to that of PhIN with 0.19 and 0.03 mT, and 1.5×10-8 s, respectively. The ratio of the average peak intensity for PhNN to the total integrated intensity was about 69% in the temperature range from 185 K to room temperature. These results are consistent with the estimation in Fig. 5. Therefore, the inclusion amount of PhNN and PhIN in 2 was determined to be x ~ 5×10-4 and y ~ 2×10-4; [(TPP)2-(PhNN)x/(PhIN)y/(N-PhMI)1.0] (x ~ 5×10-4, y ~ 2×10-4) (see also section 2.2 and 3.1). The existence of PhIN in 2 may be caused by reduction of PhNN in the synthetic process of 2 .40 We previously reported that the ESR spectra of PhNN or p-nitrophenylnitronyl nitroxide (pNPNN) radicals in CLPOT nanochannels were reproduced by a rigid-limit component of NN groups below room temperature, whereas above room temperature, they were reproduced by a single slow-region anisotropic rotational diffusion component with respect to the ESR time scale (10-9 s < τR < 10-6 s), implying that these radicals were isolated and included in the CLPOT nanochannels.38 However, the initiation temperature for spectral variation in 2 was observed at 185 K although the pore diameter of TPP nanochannels was smaller than that of CLPOT. This may be caused by weaker intermolecular interaction between the guest NN radicals and the nanochannel wall of TPP than between the NN radicals and that of CLPOT. In addition, the slightly higher initiation temperature for 2 than 1 is expected to originate from lower steric hindrance due to the smaller molecular breadth of NN than IN radicals along the molecular long axis. The spectra in the temperature range from 131 to 165 K were similar to those at 185 K


27


Page 28 of 48

despite the lower initiation temperature for rotational diffusion of PhIN radicals in the TPP nanochannels. This may be caused by the much lower inclusion amount of PhIN than PhNN in the TPP nanochannels.

3.4 Rotational diffusion activation energy for PhIN and PhNN radicals in TPP nanochannels Figure 6 shows the temperature dependence of the anisotropic rotational diffusion correlation time, τR, for PhIN in TPP nanochannels (1, cyan triangles), PhIN in TPP nanochannels with PhNN (2, yellow diamonds), PhNN in TPP nanochannels (2, red squares), and PhNN and pNPNN in CLPOT nanochannels diluted by N-PhMI38 (diagonal crosses and green circles). Here, it assumed that the molecular dynamics of PhIN and PhNN in 2 are independent of each other due to a small inclusion amount in the TPP nanochannels. The τR of each radical molecule was approximated as following the rotational Stokes-Einstein relationship and behaves as an anisotropic liquid (see section 2.4).30 The rotational activation energies for several organic radicals in 1D nanochannels, Ea, were estimated based on Arrhenius plot as shown in Table 4.38 The rotational activation energy for PhIN in the TPP nanochannels with or without co-inclusion of PhNN (2 or 1, respectively), and PhNN in the TPP nanochannels was estimated to be 19, 19 and 45 kJ mol-1, respectively, using τR values in the range of 160-260 K. The molecular orientations of NN radicals in the TPP or CLPOT nanochannels were similar: the molecular long axis is parallel to the channel axis (see section 3.3 and reference 38). Here, the magnitude of Ea of PhNN in the TPP nanochannels was larger than that of PhNN in the CLPOT nanochannels (37 kJ mol-1) as shown in Table 4. This may be caused by the smaller pore diameter of TPP


28

Page 29 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


nanochannels than that of CLPOT. The Ea of PhNN in the TPP nanochannels was smaller than that of p-NPNN in the CLPOT nanochannels (54 kJ mol-1). This is considered to be due to the smaller momentum of inertia for PhNN than that for p-NPNN. In addition, the larger Ea for PhNN than PhIN in the TPP nanochannels may be caused by the difference in the steric hindrance of NN and IN group, or by the stronger intermolecular interaction of PhNN with neighboring spacer molecules than in the case of PhIN.

Table 4. Rotational diffusion activation energies Ea of organic radicals in 1D nanochannels estimated based on Arrhenius plot. Compounds

Host

Spacers

Ea / kJ mol-1

PhNN*1

CLPOT

Non-radical

37

p-NPNN*1

CLPOT

Non-radical

54

PhNN (2; this study)

TPP

Radical and 45 non-radical

PhIN (1; this study)

TPP

Non-radical

PhIN (2; this study)

TPP

Radical and 19 non-radical

DTBN*2

TPP

Non-radical

3

TEMPO*3

TPP

Non-radical

5

TEMPONE*2

TPP

Non-radical

10

TEMPOL*4

TPP

Non-radical

26

*1

Ref. 38.

*2

Ref. 27.

*3

Ref. 23.

*4

19

Ref. 29.


29


Page 30 of 48

Figure 6 Temperature dependence of the rotational diffusion correlation time τR, for PhIN and 4-X-PhNN (X = H or NO2) in TPP or CLPOT nanochannels. Rotational activation energies, Ea, for PhIN in TPP (1), PhIN in TPP with PhNN (2), PhNN in TPP (2), PhNN in CLPOT, and p-NPNN in CLPOT38 were estimated to be 19, 19, 45, 37, and 54 kJ mol-1, respectively, using Arrhenius plot of τR values (see also Table 4).

In addition, Ea for PhIN and PhNN in the TPP nanochannels was much larger than that for a few TEMPO derivatives (5-10 kJ mol-1) such as di-t-butylnitroxide (DTBN), TEMPO and 4-oxoTEMPO (TEMPONE) radicals as shown Table 4.23,27

This is expected due to the larger

molecular size of PhIN and PhNN radicals than TEMPO. Alternatively, it could result from the stronger interaction of PhIN or PhIN with the surrounding host walls formed by phenyl rings of TPP molecules or neighboring spacers in the TPP nanochannels caused by the delocalization of


30

Page 31 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the spin density on the iminonitroxide group of IN radicals or the nitronylnitroxide group of NN radicals, as opposed to TEMPO in which the spin density is wrapped inside. On the other hand, Ea for PhIN was smaller than for 4-hydroxy-TEMPO (TEMPOL) in the TPP nanochannels (26 kJ mol-1; see Table 4).29 The Ea for TEMPOL was reported to be much larger than other TEMPO derivatives due to the different molecular orientation of TEMPOL in TPP nanochannels compared to other TEMPO derivatives: the rotational diffusion around the principal y axis of the g tensor for TEMPO derivatives other than TEMPOL,23,27 whereas the rotational diffusion around the principal z axis of the g tensor for TEMPOL.29 In PhIN and TEMPOL in the TPP nanochannels, their NO groups are approximately perpendicular to the channel axis of TPP and close to the wall of TPP nanochannels formed by the phenyl rings of TPP molecules19,29 if the rotation axis is parallel to the channel axis. The larger Ea for TEMPOL in the TPP nanochannels may be caused by the additional interaction between not NO but OH group of TEMPOL and the π-cloud on the phenyl rings of TPP wall rather than a hydrogen bridging of OH group of TEMPOL with the oxygen atoms or the lone pairs of the nitrogen atoms of the TPP structure.

4. CONCLUSION The molecular orientation and dynamics of PhIN and PhNN included in TPP nanochannels were examined using ESR spectroscopy. With N-PhMI molecules as spacers, the ESR spectra of PhIN in the TPP nanochannels were approximately simulated using a model that superimposed an anisotropic rotational diffusion component around the molecular long axis parallel to the principal y axis of the g-tensor and a rigid-limit component. The resonances were assigned to the rotational diffusion component of the PhIN radicals in the TPP nanochannels and


31


Page 32 of 48

those adsorbed on the TPP crystal, respectively. PhNN radicals were co-included in the TPP nanochannels with the PhIN radicals generated by the synthetic process and the spacers. Both radicals were found to undergo rotational diffusion above 180 K. The magnitudes of Ea for PhIN and PhNN in the TPP nanochannels were consistent with those for NN radicals in CLPOT nanochannels or those for TEMPO derivatives in TPP nanochannnels with regard to the molecular size and intermolecular interactions. These results demonstrate a new strategy for the synthesis of new organic magnets based on O1DP-ORICs. In addition, they imply that IN and/or NN group may be used to clarify the chemical and biological structure of nanomaterials such as nano-sized cavities, polymers, membranes as a new ESR spin probe technique using these substituent group in the same way as many TEMPO derivatives. Further investigations are currently underway in our laboratory.


32

Page 33 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


AUTHOR INFORMATION Corresponding Author *(H. K.) Tel/Fax: +81-555-22-6625. E-mail: [email protected]

ACKNOWLEDGMENT This work was partially supported by the Strategic Research Base Development program for Private Universities of by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT), 2009-2013. The authors wish to thank Professor T. Asaji of Nihon University for assistance with the ESR measurements, Professor T. Hashimoto of Nihon University for assistance with the TG-DTA measurements, Prof. S. Stoll of University of Washington for assistance with the EasySpin calculation, and the Organic Elemental Analysis Research Center, Kyoto University for performing the elemental analyses.


33


Page 34 of 48

REFERENCES 1. Rinkevicius, Z.; Frecus, B.; Murugan, N. A.; Vahtras, O.; Kongsted, J.; Ågren, H. Encapsulation Influence on EPR Parameters of Spin-Labels: 2,2,6,6-Tetramethyl-4methoxypiperidine-1-oxyl in Cucurbit[8]uril. J. Chem. Theory Comput 2012, 8, 257-263. 2. Kulasekharan, R.; Jayaraj, N.; Porel, M.; Choudhury, R.; Sundaresan, A. K.; Parthasarathy, A.; Ottaviani, M. F.; Jockusch, S.; Turro, N. J.; Ramamurthy, V. Guest Rotations within a Capsuleplex Probed by NMR and EPR Techniques. Langmuir 2010, 26, 6943-6953. 3. Jayaraj, N.; Porel, M.; Ottaviani, M. F.; Maddipatla, M. V. S. N.; Modeli, A.; Da Silva, J. P.; Bhogala, B. R.; Captain, B.; Jockusch, S. Turro, N. J.; Ramamurthy, V. Self Aggregation of Supramolecules of Nitroxides@Cucurbit[8]uril Revealed by EPR Spectra. Langmuir 2009,25, 13820-13832. 4. Sun, Q. F.; Iwasa J.; Ogawa, D.; Ishido, Y.; Sato, S.; Ozeki, T.; Sei, Y.; Yamaguchi, K.; Fujita, M. Self-Assembled M24L48 Polyhedra and Their Sharp Structural Switch upon Subtle Ligand Variation. Science 2010, 328, 1144-1147. 5. Fujita, D.; Ueda, Y.; Sato, S.; Mizuno, N.; Kumasaka, T.; Fujita, M. Self-Assembly of Tetravalent Goldberg Polyhedra from 144 Small Components. Nature 2016, 540, 563-566. 6. Fukino, T.; Joo, H.; Hisada, Y.; Obana, M.; Yamagishi, H.; Hikima, T.; Takata, M.; Fujita, N.; Aida, T. Manipulation of Discrete Nanostructures by Selective Modulation of Noncovalent Forces. Science 2014, 344, 499-504. 7. Nikolayenko, V. I.; Barbour, L. J.; Arauzo, A.; Campo, J.; Rawson, J. M.; Haynes, D. A. Inclusion of a Dithiadiazolyl Radical in a Seemingly Non-Porous Solid. Chem. Commun. 2017, 11310-11313.


34

Page 35 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


8. Zhou, B.; Kobayashi, A.; Kobayashi, H. Dielectric Properties of One-dimensional Water Clusters Confined in the Porous Crystal, [Co3(2,4-pyde)2(µ3-OH)2]·9H2O (2,4-pyde: pyridine-2,4-dicarboxylate). Chem. Lett. 2013, 42, 1131-1133. 9. Hertzsch, T.; Kluge, S.; Weber, E.; Budde, F.; Hulliger, J. Surface Recognition of Dipolar Molecules

Entering

Channels

of

the

Organic

Zeolite

Tris(o-

phenylenedioxy)cyclotriphosphazene. Adv. Mater. 2001, 13, 1864-1867. 10. Hertzsch, T.; Budde, F.; Weber, E.; Hulliger, J. Supramolecular-Wire Confinement of I2 Molecules in Channels of the Organic Zeolite Tris(o-phenylenedioxy)cyclotriphosphazene Angew. Chem. Int. Ed. 2002, 41, 2281-2284. 11. Kitaura, R.; Kitagawa, S.; Kubota, Y.; Kobayashi, T. C.; Kindo, K.; Mita, Y.; Matsuo, A.; Kobayashi, M.; Chang, H. -C.; Ozawa, T. C.; Suzuki, M.; Sakata, M.; Takata, M. Formation of a One-Dimensional Array of Oxygen in a Microporous Metal-Organic Solid. Science 2002, 298, 2358-2361. 12. Soegiarto, A. C.; Yan, W.; Kent, A. D.; Ward, M. D. Regulating Low-Dimensional Magnetic Behavior of Organic Radicals in Crystalline Hydrogen-Bonded Host Frameworks. J. Mater. Chem. 2011, 21, 2204-2219. 13. Cowley, H. J.; Hayward, J. J.; Pratt, D. R.; Rawson, J. M. Inclusion Chemistry of a Thiazyl Radical in Zeolite-Y. Dalton Trans. 2014, 43, 1332-1337. 14. Bardelang, D.; Giorgi, M.; Hornebecq, V.; Stepanov, A.; Hardy, M.; Rizzato, E.; Monnier, V.; Zaman, M. B.; Chan, G.; Udachin, K. Hosting Various Guests Including Fullerenes and Free Radicals in Versatile Organic Paramagnetic bTbk Open Frameworks. Cryst. Growth Des. 2014, 14, 467-476.


35


Page 36 of 48

15. Albunia, A. R.; D'Aniello, C.; Guerra, G.; Gatteschi, D.; Mannini, M.; Sorace, L. Ordering Magnetic Molecules within Nanoporous Crystalline Polymers. Chem. Mater. 2009, 21, 47504752. 16. Mon, M.; Pascual-Alvarez, A.; Grancha, T.; Cano, J.; Ferrando-Soria, J.; Lloret, F.; Gascon, J.; Pasan, J.; Armentano, D.; Pardo E. Solid-State Molecular Nanomagnet Inclusion into a Magnetic Metal–Organic Framework: Interplay of the Magnetic Properties. Chem. Eur. J. 2016, 22, 539-545. 17. Huang, J. X.; Luo, C. D.; Li, W. B.; Li, Y.; Zhang, Y. S.; Zhou, J. H.; Jiang, Q. Eccentric Magnetic Microcapsules for Orientation-Specific and Dual Stimuli-responsive Drug Release. J. Mater. Chem. B 2015, 3, 4530-4538. 18. Tateishi, K.; Negoro, M.; Nishida, S.; Kagawa, A.; Morita, Y.; M. Kitagawa, Room Temperature Hyperpolarization of Nuclear Spins in Bulk. Proc. Natl. Acad. Sci. USA 2014, 111, 7527-7530. 19. Allcock, H. R.; Siegel, L. A. Phosphonitrilic Compounds. III.1 Molecular Inclusion Compounds of Tris(o-phenylenedioxy)phosphonitrile Trimer. J. Am. Chem. Soc. 1964, 86, 5140-5144. 20. Sozzani, P.; Bracco, S.; Comotti, A.; Ferrentti, L.; Simonutti, R. Methane and Carbon Dioxide Storage in a Porous van der Waals Crystal. Angew. Chem. Int. Ed. 2005, 44, 18161820. 21. Kobayashi, H.; Ueda, T.; Miyakubo, K.; Toyoda. J.; Eguchi, T.; Tani, A. Preparation and Characterization of New Inclusion Compound with 1D Molecular Arrangement of Organic Radicals Using a One-Dimensional Organic Homogeneous Nanochannel Template. J. Mater. Chem. 2005, 15, 872-879.


36

Page 37 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


22. Kobayashi, H., Ueda, T.; Miyakubo, K.; Eguchi, T.; Tani, A. Spin-Spin Interaction of TEMPO Molecular Chains Formed in an Organic One-Dimensional Nanochannel as Studied by Electron Spin Resonance (ESR). Bull. Chem. Soc. Jpn. 2007, 80, 711-720. 23. Kobayashi, H.; Ueda, T.; Miyakubo, K.; Eguchi, T.; Tani, A. ESR Study of Molecular Dynamics and Orientation of TEMPO Included in Organic One-Dimensional Nanochannel. Phys. Chem. Chem. Phys. 2008, 10, 1263-1269. 24. Kobayashi, H.; Takamisawa, H.; Furuhashi, Y.; Nakagawa , Nakatsugawa, K.； Takeuchi, K.; Morinaga, Y. Inter-Spin Interaction of CLPOT Inclusion Compounds with 1D Molecular Chains of 4-X-TEMPO Radicals in the Temperature Range of 4.2-300 K. Bull. Chem. Soc. Jpn. 2018, 91, 375-382. 25. Barbon, A.; Zoleo, A.; Brustolon, M.; Comotti, A.; Sozzani, P. One-dimensional Clusters of 16-Doxyl-stearate Radicals in Organic Nanochannels as Studied by Electron Paramagnetic Resonance (EPR). Inorg. Chim. Acta 2008, 361, 4122-4128. 26. Süss, H. I.; Wuest, T.; Sieber, A.; Althaus, R.; Budde, F.; Lüthi, H. -P.; McManus, G. D.; Rawson, J. Hulliger, J. Alignment of Radicals into Chains by a Markov Mechanism for Polarity Formation. CrystEngComm 2002, 4, 432-439. 27. Kobayashi, H., Takeuchi, K., Asaji, T. Molecular Orientation and Dynamics of Different Sized Radicals Included in Organic 1-D Nanochannels. J. Phys. Chem. A 2013, 117, 20932101. 28. Kobayashi, H., Asaji, T., Tani, A. Preparation and Characterization of New Inclusion Compounds Using Stable Nitroxide Radicals and an Organic 1-D Nanochannel as a Template. Materials 2010, 3, 3625-3641.


37


Page 38 of 48

29. Kobayashi, H., Aoki, K.; Asaji, T. Dynamics of TEMPOL Radicals in TPP 1D Nanochannels and Different Molecular Orientation from Other TEMPO Derivatives. Chem. Lett. 2015, 44, 893-895. 30. Freed, J. H. Spin Labeling, Theory and Applications (Ed: Berliner, L. J.), Academic Press Inc., New York, US, 1976. 31. Stoll, S.; Schweiger, A. EasySpin, A Comprehensive Software Package for Spectral Simulation and Analysis in EPR. J. Magn. Reson. 2006, 178, 42-55. 32. http://www.easyspin.org/: see the manual of Chili (accessed March 13, 2018). 33. Dzikovski, B.; Tipikin, D.; Livshits, V.; Earle, K.; Freed, J. Multifrequency ESR Study of Spin-Labeled Molecules in Inclusion Compounds with Cyclodextrins. Phys Chem Chem Phys. 2009, 11, 6676-6688. 34. Aliaga, C.; Bravo-Moraga, F.; Gonzalez-Nilo, D.; Márquez, S.; Lühr, S.; Mena, G.; Rezende, M. C. Location of TEMPO Derivatives in Micelles: Subtle Effect of the Probe Orientation. Food Chem. 2016, 192, 395-401. 35. Jetti, R. K. R.; Thallapally, P. K.; Xue, F.; Mak, T. C. W.; Nangia, A. Hexagonal Nanoporous Host Structures Based on 2,4,6-Tris-4-(halo-phenoxy)-1,3,5-triazines (halo = chloro, bromo). Tetrahedron 2000, 56, 6707-6719. 36. Kobayashi, H.; Asaji, T.; Tani, A. ESR Study of the Molecular Orientation and Dynamics of Stable Organic Radicals Included in the 1-D Organic Nanochannels of 2,4,6-Tris-4(chlorophenoxy)-1,3,5-triazine. Magn. Reson. Chem. 2012, 50, 221-228. 37. Kobayashi, H.; Furuhashi, Y.; Nakagawa, H.; Asaji, T. ESR Study of Molecular Orientation and Dynamics of TEMPO Derivatives in CLPOT 1D Nanochannels. Magn. Reson. Chem. 2016, 54, 641-649.


38

Page 39 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


38. Kobayashi, H.; Morinaga, Y.; Fujimori, E.; Asaji, T. ESR Study of Molecular Orientation and Dynamics of Nitronyl Nitroxide Radicals in CLPOT 1D Nanochannels. J. Phys. Chem. A 2014, 118, 4907-4917. 39. Osiecki, J. H.; Ullman, E. F. Studies of Free Radicals. I. .alpha.-Nitronyl Nitroxides, a New Class of Stable Radicals. J. Am. Chem. Soc. 1968, 90, 1078-1079. 40. Ullman, E. F.; L. Call; Osiecki, J. H. Stable Free Radicals. VIII. New Imino, Amidino, and Carbamoyl Nitroxides. J. Org. Chem. 1970, 35, 3623-3631. 41. Hirel, C.; Vostrikova, K. E.; Pécaut, J; Ovcharenko, V. I.; Rey. P. Nitronyl and Imino Nitroxides: Improvement of Ullman’s Procedure and Report on a New Efficient Synthetic Route. Chem. Eur. J. 2001, 7, 2007-2014. 42. Nakatsuji, S.; Takai, A.; Ojima, T.; Anzai, H. Deoxygenation Reaction of Phenyl Nitronyl Nitrosides with the Strong Acceptors TCNQ4 and TCNQ. J. Chem. Res. (S) 1999, 620-621. 43. Nishida, S.; Morita, Y.; Kobayashi, T.; Fukui, K.; Ueda, A.; Sato, K.; Shiomi, D.; Takui, T.; Nakasuji, K. Spin Delocalization on Curved Surface π-system: Corannulene with Iminonitroxide. Polyhedron 2005, 24, 2200-2204. 44. Amabilino, D. B.; Veciana, J; Magnetism: molecules to materials (Ed. Miller, J. S.; Dillon, M.), part II, Molecule-Based Materials, Ch. 1, Nitroxide-Based Organic Magnets. Wiley VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2003, 1–60. 45. Dikanov, S. A.; Gulin, V. I.; Tsvetkov, Y. D.; Grigor'ev, I. A. 2 mm Electron Paramagnetic Resonance Studies of the New Types of Imidazoline Nitroxide Radicals. J. Chem. Soc. Faraday Trans. 1990, 86, 3201-3205.


39


Page 40 of 48

46. D’Anna, J. A.; Wharton, J. H. Electron Spin Resonance Spectra of α-Nitronylnitroxide Radicals; Solvent Effects; Nitrogen Hyperfine Tensor; g Anisotropy. J. Chem. Phys. 1970, 53, 4047-4052. 47. Bruce, S. D.; Higinbotham, J. Marshall, I.; Beswick, P. H. An Analytical Derivation of a Popular Approximation of the Voigt Function for Quantification of NMR Spectra. J. Magn. Reson. 2000, 142, 57-63.


40

Page 41 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC graphic


41


(a)

1

325

2

3

4

330

5

6

335

7

340

B / mT (b)

z-axis

x-axis

y-axis

325

330

335

340

B / mT


Page 42 of 48

Page 43 of 48

(e) bulk N-PhMI

(d) 3 ([TPP-(N-PhMI)])

Intensity (arb. unit)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(c) 2 ([(PhNN)/(PhIN) /(N-PhMI)]) (b) 1 ([(PhIN)/(N-PhMI)])

(a) guest free TPP

5

10

15

20

25

30

2 θ / deg



258 K

239 K

209 K

189 K

165 K

56 K

325

330

335

340

B / mT


Page 44 of 48

Page 45 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


* 253 K

* * *

*

232 K

209 K

185 K

52 K

310

315

320

325

B / mT



NN/IN IN

20/80

40/60

60/40

80/20

NN

325

330

335

340

B / mT


Page 46 of 48

Page 47 of 48

T/K

400

10-5

300

200

10-6

10-7 WR / s

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10-8 PhIN in TPP (1) PhIN in TPP (with PhNN) (2) PhNN in TPP (2) PhNN in CLPOT p-NPNN in CLPOT

10-9

10-10

2

3

4

5

103 T-1 / K-1


6

7


Page 48 of 48

(a) O O

P N O

N P

O P

O

N O

TPP (b)

O

(c)

N

N

O

N

PhNN

PhIN

(d)

O

N

N

N-PhMI


O

O

Electron Spin Resonance Study of Molecular Orientation and

Recommend Documents