Mechanoanions Produced by Mechanical Fracture of Bacterial

Sep 14, 2012 - Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan. J. Phys. Che...
0 downloads 0 Views 271KB Size
Article pubs.acs.org/JPCA

Mechanoanions Produced by Mechanical Fracture of Bacterial Cellulose: Ionic Nature of Glycosidic Linkage and Electrostatic Charging Masato Sakaguchi,*,† Masakazu Makino,† Takeshi Ohura,‡ and Tadahisa Iwata§ †

Institute for Environmental Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan Faculty of Agriculture, Meijo University, 1-501 Shiogamaguchi, Tenpaku-ku, Nagoya 468-8502, Japan § Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan ‡

ABSTRACT: Mechanoanions were produced by heterogeneous scission of the glycosidic linkages of the main chain of bacterial cellulose (BC); scission was induced by mechanical fracture of the BC in a vacuum in the dark at 77 K. The mechanoanions were detected using electron-spin-trapping methods with tetracyanoethylene. The yield of mechanoanions was positively correlated with the absolute value of the change in the Mulliken atomic charge, which was used as a descriptor of the ionic nature of the glycosidic linkage. Homogeneous scission of the glycosidic linkages induced by mechanical fracture generated mechanoradicals, the electron affinity of which was estimated on the basis of the energy of the lowest unoccupied molecular orbital for the model structure of the mechanoradical. It was concluded that the electrostatic charging of BC is caused by electron transfer from mechanoanions to mechanoradicals, which have high electron affinities. The electrostatic charge density of BC in a vacuum in the dark at 77 K was estimated to be 6.00 × 10−1 C/g.



INTRODUCTION The destruction of cellulose (e.g., threadbare cotton shirts) is a ubiquitous phenomenon that is induced by mechanical friction in the living environment. When mechanical stress is applied to cellulose, microscopic fracture of the cellulose is induced, and free radicals are produced. We reported that the mechanical fracture of bacterial cellulose1 (BC) or microcrystalline cellulose2 (MCC) in a vacuum at 77 K induced homogeneous scission of the β-1,4-glycosidic linkage comprising the cellulose main chain, resulting in a chain-end-type radical and alkoxyl radical, with pair formation. The mechanoradicals (chain-endtype and alkoxyl radicals) of BC and MCC were detected using electron spin resonance (ESR) spectroscopy and were assigned by spectral simulation using a computer program3,4 developed by one of us (M.S.). In contrast, even if an anion were produced by mechanical fracture of the β-1,4-glycosidic linkages, it could not be detected by ESR spectroscopy because the anion would not have electron spin. We reported that mechanical fracture of a synthetic polymer composed of a main chain with carbon−carbon bonds induced heterogeneous scission, resulting in mechanoanions, which were detected using an electron-spin-trapping method with tetracyanoethylene (TCNE).5−8 Although the BC main chain is not composed of carbon−carbon bonds but instead contains © 2012 American Chemical Society

glycosidic linkages that form carbon−oxygen bonds, we intended to detect the anions from mechanically fractured BC with TCNE as the TCNE anion radical (TCNE−•), using the electron-trapping method. When the BC mechanoanion comes into contact with TCNE, the TCNE withdraws an electron from the BC mechanoanion, affording TCNE−• and a chain-end-type neutral free radical of BC. Electrostatic charging of cellulose induced by friction is a rare phenomenon in the living environment. Fortunately, we found that when extremely dry BC powder was agitated by mixing in air at room temperature, electrostatic charging of BC occurred, but within several tens of seconds of ceasing the agitation, the electrostatic charging of the BC decayed. This study focused on the mechanical fracture of β-1,4glycosidic linkages of the BC main chain on the friction surface because we found the phenomenon of electrostatic charging of BC and reported1 that mechanical fracture of BC induces scission of β-1,4-glycosidic linkages. The frictional contact between two BCs should induce a microscopic fracture, that is, Received: June 26, 2012 Revised: August 22, 2012 Published: September 14, 2012 9872

dx.doi.org/10.1021/jp306261k | J. Phys. Chem. A 2012, 116, 9872−9877

The Journal of Physical Chemistry A

Article

scission of the carbon−oxygen bond constituents of the β-1,4glycosidic linkage on the friction surface. The current approach is conceptually different from conventional electrostatic charging experiments, which focus on the friction between two materials, because the mechanical fracture of the BC is executed in vacuum at 77 K in the dark. This experimental technique is superior to conventional experimental techniques because the actual contact area and surface geometry do not need to be known, and the environmental conditions can be controlled to prevent the decay of the primary products or generation of secondary reactions. Herein it is shown that BC mechanoanions were detected as a result of mechanical fracture of BC in a vacuum at 77 K in the dark, and the electron affinity of the BC mechanoradical was estimated on the basis of the energy level of the lowest unoccupied molecular orbital (LUMO) for the model structure of the mechanoradical. It was concluded that the electrostatic charging is the result of electron transfer from BC mechanoanions to BC mechanoradicals, which have high electron affinities. The electrostatic charge density of BC in a vacuum in the dark at 77 K was calculated.

measurements. Surface areas were calculated using the Brunauer−Emmett−Teller (BET) method in the relative pressure (P/P0) range from 0.13 to 0.27.



RESULTS AND DISCUSSION The solid line in Figure 1a shows the observed ESR spectrum of BC fractured with TCNE in a vacuum in the dark at 77 K.



EXPERIMENTAL SECTION Materials. Gluconacetobacter xylinus (=Acetobacter xylinum) JCM9730 was the strain used for the production of BC. The culture conditions and the purification of BC followed the procedure described in a previous article.1 Mechanical Fracture. BC was dried in vacuum (0.6 Pa) at 373 K for 7 h. TCNE (Wako Chemicals) was purified by sublimation in vacuum. The predried BC (0.421 g) and TCNE (0.048 g) were introduced into a glass ball mill. The glass ball mill containing BC and TCNE was evacuated at 273 K for 2 h under a pressure of 0.6 Pa, sealed off, and placed in a Dewar flask filled with liquid nitrogen. The sample was mechanically fractured using a homemade vibration ball-mill apparatus9 for 7 h at 77 K in a vacuum in the dark. After the sample was milled, the powder sample was dropped into the ESR sample tube attached to the top of the glass ball mill by turning it over in the liquid nitrogen in the dark. ESR Measurements. ESR spectra were observed at a microwave power level of 2 μW (to avoid power saturation) and with 100 kHz field modulation using a Bruker EMX Plus spectrometer (X-band) equipped with a helium cryostat (Oxford ESR 900) and a temperature controller (Oxford ITC4). ESR Spectral Simulations. ESR spectral simulations were carried out using a computer program3,4 developed by one of us (M.S.) to calculate the line-shape equation of the ESR spectrum in the solid state, with an anisotropic g tensor and hyperfine splitting tensor A. Visible-Light Irradiation. The fractured sample in the ESR sample tube was irradiated in the dark in a vacuum at 77 K using a visible-light source (Cold Light HL100E, HOYASCHOTT) with a glass filter (Toshiba IRP 70 glass filter; 400 nm < λ < 800 nm). Calculation of Mulliken Atomic Charges and LUMO Energies. The Mulliken charge of each atom and the LUMO energy for the model structure of BC and the model structure of the mechanoradical were calculated using the Gaussian 09 software package.10 BET Measurements. The nitrogen adsorption isotherm at 77 K was obtained using a Quantachrome Autosorb-1 system. All samples were degassed at 298 K for over 5 h before

Figure 1. (a) Solid line (): ESR spectrum of fractured BC with TCNE in a vacuum in the dark at 77 K. Dashed line (---): simulated ESR spectrum of BC mechanoradicals. (b) Solid line (): ESR spectrum of fractured BC with TCNE in a vacuum in the dark at 77 K before visible-light irradiation. Dashed line (---): after visible-light irradiation for 45 min. (c) Solid line (): Incremental spectrum obtained by subtracting the spectrum before visible-light irradiation from that after visible-light irradiation for 45 min. Dashed line (---): simulated ESR spectrum of TCNE−•.

The outer regions of the observed spectrum (indicated by arrows) are ascribed to BC mechanoradicals resulting from the mechanical fracture of BC.1,2 The simulated spectrum of the BC mechanoradical is shown as the dashed line in Figure 1a. The spectrum of the center region of the observed spectrum is similar to that of TCNE−• in the case of synthetic polymers.5−8 The intensity of the center peak of the spectrum was increased by visible-light irradiation (dashed line in Figure 1b); however, the outer regions were unaltered. The solid line in Figure 1c shows the incremental spectrum obtained by subtracting the spectrum of the fractured BC in the dark (solid line in Figure 1b) from the spectrum obtained following visible-light irradiation for 45 min (dashed line in Figure 1b). Yoshida and co-workers11 reported the ESR parameters of TCNE−•: giso = 2.00277 and AN,iso = 0.1571 mT. In our study, the simulated spectrum of TCNE−• (dashed line in Figure 1c) was obtained using the following ESR parameters: (gxx, gyy, gzz) = (2.0021 ± 0.0001, 2.0031 ± 0.0001, 2.0031 ± 0.0001), giso = 2.0028, (AN,xx, AN,yy, AN,zz) = (0.54 ± 0.01, −0.06 ± 0.01, −0.06 ± 0.01 mT), and AN,iso = 0.14 mT. The peak positions of the simulated spectrum were identical to those of the incremental spectrum (solid line in Figure 1c), and the profile of the simulated spectrum was almost identical to that of the 9873

dx.doi.org/10.1021/jp306261k | J. Phys. Chem. A 2012, 116, 9872−9877

The Journal of Physical Chemistry A

Article

Figure 2. Homogeneous scission of the β-1,4-glycosidic linkage of BC produces mechanoradicals: Ia and Ib at scission I and IIa and IIb at scission II. Heterogeneous scission of the β-1,4-glycosidic linkage of BC produces mechanoions: cation Ia+ and anion Ib− at scission I and anion IIa− and cation IIb+ at scission II.

Cell−C1−O−C2−Cell′ → Cell−C1−O • + •C2−Cell′

incremental spectrum. Therefore, the incremental spectrum was identified as TCNE−•. Furthermore, no ESR signal was detected from unfractured BC with TCNE exposed to visiblelight irradiation. This result indicates that chemical bonds were not destroyed by the visible-light irradiation. The results mentioned above strongly suggest that when the BC mechanoanion comes into contact with TCNE during milling, the TCNE withdraws an electron from the BC mechanoanion, affording TCNE−•. After milling, the BC mechanoanion releases an electron upon visible-light irradiation, and TCNE traps the electron to produce TCNE−•. A scheme showing the homogeneous and heterogeneous scissions of the β-1,4-glycosidic linkages of BC and the resulting products is presented in Figure 2. For the homogeneous scission, we reported previously1 that mechanical fracture of BC at scission I induces pair formation of the radicals Ia and Ib, and fracture at scission II forms IIa and IIb. In the heterogeneous scission, we assumed the pair formation of Ia+ (−C1+) and Ib− (−C2−O−) and of IIa− (−C1−O−) and IIb+ (−C2+), as shown in Figure 2. Furthermore, since oxygen atoms have a higher electron affinity than carbon atoms, we neglected the formation of +O−C2− and −C1−O+. The reaction scheme is described in the following sections. (i). During Milling in a Vacuum in the Dark at 77 K. Homogeneous scission of the β-1,4-glycosidic linkages at scission I or scission II (Figure 2) produces mechanoradicals Ia and Ib or IIa and IIb with pair formation:

IIa:[1R 1]

(1b) 1

1

where [ R1] and [ R2] in eq 1a are the radical concentrations of Ia and Ib, respectively, and [1R1] and [1R2] in eq 1b are those of IIa and IIb, respectively. [1R1] = [1R2] as a result of pair formation of mechanoradicals and equal scission probabilities at scission I and scission II in the homogeneous scission, on the basis of the analysis of the ESR spectra presented in our previous paper.1 Simultaneously, heterogeneous scission of the β-1,4-glycosidic linkage at scission I or scission II produces mechanoions Ia+ and Ib− or IIa− and IIb+, respectively, with the assumption of pair formation of mechanoions and equal scission probabilities at scission I and scission II in the heterogeneous scission: Cell−C1−O−C2−Cell′ → Cell−C1+ + −O−C2−Cell′ Ia +

Ib−

Cell−C1−O−C2−Cell′ → Cell−C1−O− + +C2−Cell′ IIa −

IIb+

8

In our previous paper on synthetic polymers, a tendency for the electron affinity of mechanoradicals was suggested on the basis of the electron affinity of the substituents attached to the carbon−carbon bond of the polymer main chain. However, the speculation was qualitative and limited to carbon−carbon bonds. Therefore, this speculation cannot be applied to BC mechanoradicals. In this study, we assumed that the stability of the anion formed by the capture of an electron is closely related to the electron affinity of the mechanoradical. Thus, we calculated the LUMO energies of Ib-m and IIa-m (Figure 3) as model

Cell−C1−O−C2−Cell′ → Cell−C1 • + •O−C2−Cell′ Ia:[1R 1]

IIb:[1R 2]

Ib:[1R 2]

(1a) 9874

dx.doi.org/10.1021/jp306261k | J. Phys. Chem. A 2012, 116, 9872−9877

The Journal of Physical Chemistry A

Article visible‐light irradiation

Cell−C1−O− ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ Cell−C1−O • + e− IIa



IIa:[3R 1]

−• e− + TCNE → TCNE 3

(4b)

[ T1]

3

(4a)

3

where [ R2] and [ R1] are the radical concentrations of Ib and IIa in eqs 3a and 4a, respectively, and [3T2] and [3T1] are those of TCNE−• in eqs 3b and 4b, respectively. We assumed that [3R2] = [3T2] and [3R1] = [3T1] on the basis of eqs 3a, 3b, 4a, and 4b and therefore that [3R2] = [3T2] = [3R1] = [3T1] assuming equal scission probabilities at scission I and scission II in the heterogeneous scission. The ratio of the total radical concentrations after and before visible-light irradiation, IA, is given by the ratio of the integrations of the total ESR spectra after and before visiblelight irradiation:

Figure 3. Model structures of BC mechanoradicals for LUMO energy calculations: Ib-m for Ib and IIa-m for IIa.

structures of mechanoradicals Ib and IIa based on the productions shown in Figure 2; the molecular structures of the models were optimized at the unrestricted B3LYP/6311G(d,p) level using the Gaussian 09 software package.10 We used the LUMO energy as an effective descriptor to estimate the electron affinity. The LUMO energies of Ib-m (for Ib) and IIa-m (for IIa) were −0.1585 au and −0.1604 au, respectively. The LUMO energy of TCNE was −0.1875 au [restricted B3LYP/6-311G(d,p)]. Thus, the LUMO energy can be regarded as a useful descriptor for estimating the electron affinity. TCNE has a high electron affinity relative to the model compounds for the BC mechanoanions. Therefore, TCNE can easily draw out an electron from the mechanoanions to form TCNE−•. An electron capture reaction similar to that of TCNE would be expected for Ia+ and/or IIb+. However, these reactions were neglected in our analysis because the calculated structures of Ia+ and IIb+ differed completely from the input structures; for example, during the optimization, some cleavage of small regions occurred. Furthermore, the concentrations of both Ia+ and IIb+ were very low in comparison to that of TCNE. During milling, when mechanoanion Ib− or IIa− comes into contact with TCNE (the contact is promoted by physical mixing in the vibration glass ball mill during milling), electron transfer from Ib− or IIa− to TCNE in the dark at 77 K produces TCNE−•, and the mechanoanion is transformed into a neutral free radical:

IA =

[1R 1] + [2 R 1] + [3R 1] [1R 1] + [2 R 1]

The ratio of the TCNE−• concentrations after and before visible-light irradiation, IB, is given by the ratio of the heights of the center peaks from TCNE−•: IB =

[2 T1] + [3T1] 2

[ T1]

=

[2 R 1] + [3R 1] [2 R 1]

Figure 4 shows that both IA and IB increase with increasing duration of visible-light irradiation and then ultimately reach



−• O−C2−Cell′ + TCNE → •O−C2−Cell′ + TCNE 2 Ib−

Ib:[2 R 2]

Figure 4. Effect of visible-light irradiation on IA, the ratio of the total radical concentrations after and before visible-light irradiation (○), and on IB, the ratio of the TCNE−• concentrations after and before irradiation (●).

[ T 2]

(2a) −

−•

Cell−C1−O + TCNE → Cell−C1−O • + TCNE 2 IIa −

IIa:[2 R 1]

[ T1]

(2b)

plateaus. If it is assumed that all of the mechanoanions react with TCNE after the visible-light irradiation, the anionic yield Fan (i.e., the ratio of heterogeneous scission to total scission) can be expressed as follows:

where [2R2] and [2T2] are the radical concentrations of Ib and TCNE−•, respectively, in eq 2a and [2R1] and [2T1] are those of IIa and TCNE−•, respectively, in eq 2b. We assumed that [2R2] = [2T2] = [2R1] = [2T1] on the basis of equal scission probabilities at scission I and scission II in the heterogeneous scission. (ii). Visible-Light Irradiation after Milling. After milling, the fractured sample in a vacuum at 77 K in the dark was exposed to visible-light irradiation at 77 K. The mechanoanion (Ib− or IIa−) releases an electron as a result of the visible-light irradiation, resulting in a neutral free radical (Ib or IIa); TCNE then traps the electron to form TCNE−•:

Fan = =

O−C2−Cell′ ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ •O−C2−Cell′ + e− −

Ib:[3R 2]

Ib

−• e− + TCNE → TCNE 3 [ T 2]

[1R 1] + [2 R 1] + [3R 1] IB(IA − 1) IA(IB − 1)

The value of Fan for BC was calculated as 0.684 ± 0.005 using the plateau values IA = 1.218 ± 0.004 and IB = 1.356 ± 0.005 for a 45 min duration of visible-light irradiation. It seems quite plausible that Fan should depend on the ionic nature of the β-1,4-glycosidic linkage. In our previous paper on synthetic polymers,8 the ionic nature of a carbon−carbon bond consisting of a polymer main chain was deduced qualitatively on the basis of the binding energies of the substituents attached

visible‐light irradiation



[2 R 1] + [3R 1]

(3a) (3b) 9875

dx.doi.org/10.1021/jp306261k | J. Phys. Chem. A 2012, 116, 9872−9877

The Journal of Physical Chemistry A

Article

mechanoanions under visible-light irradiation at room temperature under humid conditions. Cellulose is located at a nearly neutral position in the triboelectric series,15 polyethylene is located at a negative position, and polytetrafluoroethylene (PTFE) is in the most negative position. In contrast, the average LUMO energy of BC [−0.1595 au, the average of the values for Ib-m (−0.1585 au) and IIa-m (−0.1604 au)] in a vacuum in the dark at 77 K is lower than that of PTFE (−0.1188 au). This result indicates that the position of BC in the triboelectric series should be more negative than that of PTFE. The sign of the charge can be assumed on the basis of the LUMO energy of the mechanoradical (i.e., from the chemical structure). In conventional environments, one can deduce that BC mechanoanions must release electrons in a significant quantity under conventional friction in a humid atmosphere at room temperature under visible-light irradiation, resulting in a nearly neutral position in the triboelectric series.

to the carbon−carbon bond. However, this was qualitative speculation and was limited to carbon−carbon bonds. To evaluate the ionic nature of the β-1,4-glycosidic linkage, we calculated the difference between the Mulliken atomic charges of two adjacent atoms (i.e., 1C−O or O−2C) in the β1,4-glycosidic linkages 1C−O−2C of which the BC main chain is composed, and then we applied their absolute values to the evaluation. These descriptors, which are denoted as |Δ(1C−O)| or |Δ(O−2C)| hereafter, clearly reflect the ionic nature of the bonds. We calculated |Δ(1C−O)| and |Δ(O−2C)| for the model structure illustrated in Figure 5 and obtained the values

Figure 5. Model structure of a β-1,4-glycosidic linkage of BC used for the calculation of the absolute value of the change in Mulliken atomic charge.



CONCLUSION Mechanical fracture of BC produces mechanoradicals by homogeneous scission of the β-1,4-glycosidic linkage of BC and mechanoanions by heterogeneous scission of the linkage. The mechanoanions were detected as TCNE−• using the electron-trapping method with TCNE and were identified by spectral simulations. The anionic yield of BC by mechanical fracture was 0.684 ± 0.005 after 45 min of exposure to visiblelight irradiation. The electron affinities of TCNE and the BC mechanoradicals were deduced from calculations of the LUMO energies of TCNE and model structures of the mechanoradicals. TCNE had the lowest LUMO energy among the mechanoradicals. Therefore, TCNE could easily capture an electron from the mechanoanion, resulting in TCNE−•. We understand electrostatic charging as a mechanical scission of the chemical bond of BC at the friction surface. We propose that the electrostatic charging of the BC is induced by electron transfer from the mechanoanions to the mechanoradicals, which have high electron affinities. The electrostatic charge density of BC in a vacuum in the dark at 77 K was calculated to be 6.00 × 10−1 C/g or 5.00 × 10−1 C/m2. The charge can be speculated to be negative on the basis of the chemical structure (i.e., the LUMO energy) of the BC mechanoradicals.

1.03754 and 0.76347, respectively. In view of the value of Fan (0.684 ± 0.005), it should be noted that the values of both descriptors |Δ(1C−O)| (or |Δ(O−2C)|) and Fan were relatively large. In contrast, polyethylene (PE), with a calculated value of |Δ(1C−2C)| = 0.00, had small value of Fan (0.11 ± 0.10).8 It seems to be quite plausible that Fan depends on the ionic nature of the linkage. A generation mechanism for electrostatic charging on the friction surface of celluloses A and B can be deduced as follows. When the cellulose mechanoanion A− (Ib− or IIa−) produced by heterogeneous scission at the friction surface comes into contact with the cellulose mechanoradical B· produced by homogeneous scission at the friction surface, B· draws the electron away from A−, affording the anion B− and the neutral free radical A·. The electron transfer from A− to B· as a result of contact and separation induces a positive charge on the nanodomain of cellulose A and a negative charge in the nanodomain of cellulose B. This phenomenon should form a random mosaic pattern with oppositely signed charges at the nanodomain level on the surface. This idea is supported by the reports of Grzybowski and co-workers,12,13 in which a random mosaic of oppositely charged regions of nanoscopic dimensions was induced by contact between polydimethylsiloxane plates. Unfortunately, the charge generation mechanism was not explained clearly. The electrostatic charging of BC was estimated as follows: The concentration of BC mechanoanions (Ib− and IIa−) was 3.74 × 1018 anions/g, using Fan = 0.684 and a BC radical concentrationof (1.73 × 1018 spins/g. The amount of charge generated by friction was estimated to be 6.00 × 10−1 C/g, assuming each anion had one electron (1.602 × 10−19 C).14 The specific surface area (12.0 m2/g) was obtained using the BET method. The amount of charge generated by friction can be estimated at 5.00 × 10−1 C/m2. This estimated value is extremely high in comparison with the value obtained in conventional cellulose experiments. In conventional experiments, the amount of charge from cellulose is observed in a humid atmosphere and under visible-light irradiation at room temperature. Therefore, we surmise that the extremely low value from conventional cellulose experiments is the result of almost complete decay of the cellulose



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This research was partially supported by the Ministry of Education, Science, Sports and Culture through a Grant-in-Aid for Scientific Research (C) (21580208) and a Grant-in-Aid for Challenging Exploratory Research (24658157). Thanks are due to the Instrument Center of the Institute for Molecular Science for assistance in obtaining the low-temperature ESR spectra.



REFERENCES

(1) Sakaguchi, M.; Ohura, T.; Iwata, T.; Takahashi, S.; Akai, S.; Kan, T.; Murai, H.; Fujiwara, M.; Watanabe, O.; Narita, M. Biomacromolecules 2010, 11, 3059−3066. (2) Sakaguchi, M.; Ohura, T.; Iwata, T.; Takahashi, S; EnomotoRogers, Y. Polym. Degrad. Stab. 2012, 97, 257−263.

9876

dx.doi.org/10.1021/jp306261k | J. Phys. Chem. A 2012, 116, 9872−9877

The Journal of Physical Chemistry A

Article

(3) Sakaguchi, M.; Yamamoto, K.; Miwa, Y.; Shimada, S.; Sakai, M.; Iwamura, T. Macromolecules 2007, 40, 1708−1712. (4) Sakaguchi, M.; Iwamura, T.; Yamamoto, K.; Miwa, Y.; Shimada, S.; Sakai, M. Macromolecules 2008, 41, 253−257. (5) Sakaguchi, M.; Kinpara, H.; Hori, Y.; Shimada, S.; Kashiwabara, H. Polymer 1984, 25, 944−946. (6) Sakaguchi, M.; Kinpara, H.; Hori, Y.; Shimada, S.; Kashiwabara, H. J. Polym. Sci., Part B: Polym. Phys. 1987, 25, 1431−1437. (7) Sakaguchi, M. In Polymeric Materials Encyclopedia; Salamone, J. C., Ed.; CRC Press: Boca Raton, FL, 1996; Vol. 6 (M−O), pp 4048− 4051. (8) Sakaguchi, M.; Miwa, Y.; Hara, S.; Sugino, Y.; Yamamoto, K.; Shimada, S. J. Electrost. 2004, 62, 35−49. (9) Sakaguchi, M.; Sohma, J. J. Polym. Sci., Polym. Phys. Ed. 1975, 13, 1233−1245. (10) Frisch, M. J.; et al. Gaussian 09, revision A.1; Gaussian, Inc.: Wallingford, CT, 2009. (11) Yoshida, J.-i.; Tamao, K.; Kumada, M. J. Am. Chem. Soc. 1980, 102, 3269−3270. (12) Baytekin, H. T.; Patashinski, A. Z.; Branicki, M.; Baytekin, B.; Soh, S.; Grzybowski, B. A. Science 2011, 333, 308−312. (13) Apodaca, M.; Wesson, P. J.; Bishop, K. J. M.; Ratner, M. A.; Grzybowski, B. A. Angew. Chem., Int. Ed. 2010, 49, 946−949. (14) CRC Handbook of Chemistry and Physics, 76th ed.; Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 1995−1996; p 1. (15) Henniker, J. Nature 1962, 196, 474.

9877

dx.doi.org/10.1021/jp306261k | J. Phys. Chem. A 2012, 116, 9872−9877