In this study, the electrical conducting properties of six types of biomaterials, comprising cellulose and chitosan derived from terrestrial plants and marine products, respectively, were investigated using electron spin resonance (ESR) and Schottky junction characteristics. Kenaf, chitosan, conifer, and RCH2OH (R = C11H17O9) exhibited ESR spectra showing unpaired electrons at 295 K, demonstrating rectifying effects at room temperature. In contrast, RCOONa (C12H17O11Na) and α-chitin, which did not exhibit observable ESR spectra, showed ohmic conduction behavior. The ESR g value was used to determine the organic radical species, suggesting that electrons originate from the glycosidic C1–O1–C4 radical in cellulose and the aminyl N–H radical in chitosan. RCOONa and α-chitin, which possess C=O bonds, suppress electron-induced effects and consequently inhibit the transport of free radicals, resulting in ohmic conduction.

Highly durable, lightweight, and degradable celluloses, which are the most abundant organic compounds in nature, have recently attracted considerable attention as eco-friendly and renewable biomaterials owing to their thermal stability, robustness, and low weight.1–4 Chitosan, primarily derived from marine products, is the second most abundant biomass after cellulose and has a structure similar to that of cellulose.5,6 Biodegradable chitosan has gained significant interest owing to its remarkable properties,7 including filtration capabilities.8 

We previously reported the development of supercapacitors from 2, 2, 6, 6-tetramethylpiperidine-1-oxyl radical (TEMPO)-oxidized cellulose nanofibers (CNFs)9–11 and n-type semiconductors from kenaf cellulose nanofibers (KCNFs)12 and chitosan nanofibers (ChNFs)13 fabricated using mechanical defibration methods. Hall effect measurements revealed n-type semiconductors with a carrier concentration of 9.89 × 1015/cm3 and a mobility of 10.66 cm2/Vs.14 However, unpaired electrons available for semiconductors have not yet been confirmed in other terrestrial plants and marine products. In this study, we employed electron spin resonance (ESR) to investigate whether electrons capable of acting as semiconductors are present in six representative types of CNFs and ChNFs derived from these natural products. ESR is a useful method for identifying radicals in organic materials.15,16 In biochemistry, it has been used to detect unpaired electrons in metal clusters bound by proteins involved in electron conduction.17,18 However, ESR studies of bioconductors are less common than those of polymeric conductors. Furthermore, limited research has been conducted on the electronic applications of biomaterials sourced from terrestrial plants or marine products.

Conifer CNFs (13A21-nanoforest-S-1NBC) with diameters ranging from 20 to 60 nm were mechanically prepared by Chuetsu Pulp Industry. Kenaf CNFs with a diameter of 11 nm were mechanically miniaturized by Tohoku University.14 RCH2OH was produced by removing xanthate groups from cellulose molecules (RCNF, 3–8 nm in diameter, WPA240326F1, Rengo Industry, Japan).19 NaOH extracts of TEMPO-oxidized celluloses (RCOONa) with diameters of 3 nm were prepared according to a previous method.20 α-Chitin and chitosan ChNFs were processed at 400 rpm for 5 s using a 2% (w/v) water dispersion (BiNFi-s chitin, SFo-20002; BiNFi-s; chitosan, EFo-08002; Sugino Machine, Japan).13 

ESR measurements were conducted in air using 0.2 g of sample placed in a quartz tube and an X-band ESR spectrometer (JES-X330, JEOL) [power: 2 mW; modulation width: 0.5 mT; time constant: 0.1 s; sweep time: 60 s] at 295 and 103 K. The g values were determined relative to the third and fourth signals from the lower magnetic field (g = 1.981) of Mn2+ in MgO. Current–voltage (I–V) characteristics, based on a Schottky coupling between carbon and the CNF specimen,14 were measured within 30 min after sample preparation under DC voltages ranging from −200 to +200 V in air at a sweep rate of 1.24 V/s using a Precision Source/Measure Unit (B2911A, Agilent). AC impedance was measured using a potentiostat/galvanostat (SP-150, BioLogic Science). All electronic measurements were performed in an Al shield box to prevent electromagnetic interference from the surroundings. The local DOS of oxygen (c) and local potential Fermi energy at positions along the a-axis (fractional coordination) for C1OC4, C6COONa, and C6OC1 of C12H20O10 and C12H17O11Na were determined using plane-wave-based first-principles density functional calculations (VASP 5.3).21 

ESR measurements,15,16 which are the only method for observing radicals in organic materials,17,18 were conducted at 295 and 103 K to investigate the origin of electrons in the six types of CNFs and ChNFs. The ESR spectra at 295 and 103 K for the four terrestrial plants and two marine products are shown in Figs. 1(a) and 1(b), respectively. All the spectra, except for those of RCOONa and α-chitin, exhibit singlet symmetrical peaks. At 295 K, the peak intensities were highest for kenaf, followed by chitosan, conifer, and RCH2OH, while RCOONa and α-chitin exhibited smaller peaks. In contrast, at 103 K, kenaf showed a prominent peak, while the other materials had smaller peaks. The higher radical peak of kenaf can likely be attributed to its harder structure compared with those of the other mechanically defibrated materials and its finer diameter of 11 nm.22 In general, the ESR peaks of artificial materials are higher at lower temperatures than at room temperature.23 However, the ESR peaks of biomaterials are notably weaker, except for kenaf and chitosan, which may be related to the liquid–solid transformation of bound water.

FIG. 1.

ESR spectra of plant CNPs (conifer, kenaf, RCH2OH, and RCOONa) and marine product ChNFs (α-chitin and chitosan) at (a) 295 K and (b) 103 K.

FIG. 1.

ESR spectra of plant CNPs (conifer, kenaf, RCH2OH, and RCOONa) and marine product ChNFs (α-chitin and chitosan) at (a) 295 K and (b) 103 K.

Close modal

Because unpaired electrons observed in the ESR spectrum are highly sensitive to the surrounding molecular arrangement, the g value, which is defined as the magnetic field value at which the ESR signal curve intersects zero intensity, serves as a critical guide for identifying the organic radical species.24 Based on the molecular structure of cellulose (C6H10O5)n, the g values of 2.004 56 and 2.004 70 at 295 K are higher than the free electron spin value g = 2.002325 but lower than the values for lignin (g = 2.0071–2.0073).22 Thus, these results suggest that the electrons from kenaf, conifer, and RCH2OH originate from the C1–O1–C4 radicals. The 295 K peak of chitosan exhibits a singlet symmetrical ESR signal centered at g = 2.005 80, which is within the range of 2.0054–2.0059 reported for chitosan.26,27 Moreover, this g value is consistent with that expected for the aminyl radical, N–H, found in amorphous chitosan,28 as reported by Saiki et al.29 and Gryczka et al.30 The results, including the g values, ESR signal intensities, and radical groups at 295 and 103 K, are summarized in Table I. Table I clearly shows that the amount of unpaired electron radicals varied depending on the type of terrestrial plants and marine products. However, the organic radical species in RCOONa could not be identified. The degree of deacetylation may affect chitosan, which will be discussed in the subsequent report.

TABLE I.

g values and ESR signal its intensities at 295 and 103 K, along with radical groups for the six types of specimens. ESR signal intensity: ◎ (large) > ○ (medium) > □ (small) > △ (extremely small).

295 K103 K
Speciesg-valueIntensityg-valueIntensityRadical group
Conifer 2.004 59 □ 2.005 55 △ C1–O1-O4 
Kenaf 2.004 56 ◎ 2.004 57 ○ C1–O1–O4 
Pure R 2.004 70 △ 2.004 90 △ C1–O1–O4 
RCOONa 2.005 65 △ 2.005 85 △ 
α-chitin 2.005 39 △ 2.005 48 △ N–H 
Chitosan 2.005 80 ○ 2.005 64 △ N–H 
295 K103 K
Speciesg-valueIntensityg-valueIntensityRadical group
Conifer 2.004 59 □ 2.005 55 △ C1–O1-O4 
Kenaf 2.004 56 ◎ 2.004 57 ○ C1–O1–O4 
Pure R 2.004 70 △ 2.004 90 △ C1–O1–O4 
RCOONa 2.005 65 △ 2.005 85 △ 
α-chitin 2.005 39 △ 2.005 48 △ N–H 
Chitosan 2.005 80 ○ 2.005 64 △ N–H 

When a semiconductor is bonded to a metal with a high work function, a depletion layer known as a Schottky barrier is created. Electricity flows from the metal to the semiconductor if the semiconductor is n-type and from the semiconductor to the metal if it is p-type. Figure 2 shows the I–V characteristics at 298 K for the six types of biospecimens bonded to carbon with a work function of 5. The conifer (a), kenaf (b), RCH2OH (c), and chitosan (f) samples with a work function of 4.7710 exhibited rectifying effects (black lines), whereas RCOONa (d) and α-chitin (e) showed ohmic linear behavior (blue lines) with low current levels below 30 μA. Comparing these results with the number of radical electrons, we found that the former group of materials, which exhibit large, medium, and small ESR peaks, are n-type semiconductors. In contrast, the latter group with weak ESR peaks consists of ohmic materials as well as pure intrinsic Si without radical electrons, despite RCOONa having a work function of 2.834.11 Based on these I–V characteristics and ESR g values, the six samples can be classified into two types, semiconductors and ohmic conductors, as shown in Fig. 3. The C1O1C4 and NH radicals were divided into two groups. However, the origin of the radical in RCOONa could not be identified based on its g value.

FIG. 2.

IV characteristics of terrestrial plant CNFs [(a) conifer, (b) kenaf, (c) RCH2OH, and (d) RCOONa] and marine product ChNFs [(e) α-chitin and (f) chitosan].

FIG. 2.

IV characteristics of terrestrial plant CNFs [(a) conifer, (b) kenaf, (c) RCH2OH, and (d) RCOONa] and marine product ChNFs [(e) α-chitin and (f) chitosan].

Close modal
FIG. 3.

Diagram showing the six samples plotted by maximum current values from IV characteristics and g values from ESR. The samples are classified into two types: semiconductors and ohmic conductors.

FIG. 3.

Diagram showing the six samples plotted by maximum current values from IV characteristics and g values from ESR. The samples are classified into two types: semiconductors and ohmic conductors.

Close modal

Although we did not examine I–V characteristics at lower temperatures, terrestrial plants and marine products, with the exception of kenaf and chitosan, cannot be used as semiconductors at low temperatures.

To non-destructively evaluate the electrical properties of these specimens, we measured the AC impedance from 1 to 1 MHz using Nyquist and Bode diagrams at 298 K. Conifer served as a semiconductor, while RCOONa was used as a normal conductor. The Nyquist and Bode diagrams and frequency-dependent capacitances, Cp and Cs, for conifer are shown in Figs. 4(a)4(c), respectively. The impedance variation of conifer with frequency shows two semicircles, corresponding to lower and higher resistances (R1 and R2; R1 < R2) and capacitances (C1 and C2; C1 < C2) based on its semiconducting behavior.12 The second large semicircle, characteristic of Debye-type relaxation, reveals a relaxation time of 0.024 s at a peak frequency fmax of 6.75 Hz, indicating dielectric dispersion caused by interfacial polarization.31 Although both series (Cs) and parallel (Cp) capacitances increase parabolically with frequency, Cs increases more significantly than Cp, reaching a value five orders of magnitude larger at 1 mHz.

FIG. 4.

Nyquist plots for (a) conifer and (d) RCOONa. Frequency dependencies of real and imaginary impedances for (b) conifer and (e) RCOONa. Capacitance behavior as a function of frequency for (c) conifer and (f) RCOONa.

FIG. 4.

Nyquist plots for (a) conifer and (d) RCOONa. Frequency dependencies of real and imaginary impedances for (b) conifer and (e) RCOONa. Capacitance behavior as a function of frequency for (c) conifer and (f) RCOONa.

Close modal

In contrast, the Nyquist diagram of RCOONa shows a half semicircle followed by a vertical straight line, which is often associated with electric storage behavior.9–11 Both the real and imaginary impedances increase with decreasing frequency; however, the latter surpasses the former at 1 mHz. Although Cp and Cs increase parabolically with decreasing frequency, the Cs of RCOONa is five orders of magnitude smaller than that of conifer.

The Nyquist diagram for conifer [Fig. 4(a)] suggests that the electrical circuits of the four types of samples exhibiting semiconducting characteristics are electrically lumped constant circuits, as shown in Fig. 5(a). Figure 5(b) shows the image of a semiconductor composed of the estimated conifer CNFs and their boundaries, measured using the direct-current method. Rfb and Rer represent the resistances of the boundaries between the nanofibrils and the electrode interface resistance, respectively. The large capacitance observed in Fig. 4(c) is likely related to the substantial capacitances induced by the electric double layer of pseudo-solid water formed at the interface between the cellulose molecules and the electrode,32,33 as shown in Fig. 5(d). Sasaki and Shinyashiki reported that hydrated water at the electrode interface transitions into a pseudo-solid phase with an electric double layer in a narrow confinement of 0.311 nm,34 similar to that of an electrolyte. In contrast, the Nyquist diagram for RCOONa [Fig. 4(c)] shows an electrically distributed circuit, as shown in Fig. 5(c).

FIG. 5.

(a) Equivalent circuit corresponding to the Nyquist diagram shown in Fig. 4(a) for conifer. (b) Schematic of a semiconductor composed of conifer CNFs and their boundaries, measured using the direct current method. (c) Electrically distributed circuit corresponding to the Nyquist diagram shown in Fig. 4(d) for RCOONa. (d) Schematic of Rer for the electric double layer.

FIG. 5.

(a) Equivalent circuit corresponding to the Nyquist diagram shown in Fig. 4(a) for conifer. (b) Schematic of a semiconductor composed of conifer CNFs and their boundaries, measured using the direct current method. (c) Electrically distributed circuit corresponding to the Nyquist diagram shown in Fig. 4(d) for RCOONa. (d) Schematic of Rer for the electric double layer.

Close modal

We further investigated the plant materials, specifically examining why conifer, kenaf, and RCH2OH exhibit semiconducting characteristics, while RCOONa shows ohmic conduction behavior, despite their electronic origins being identified as C1–O1–C4 through ESR measurements. This difference may be related to the presence or absence of the C=O functional groups, as shown in Figs. 6(a) and 6(b), respectively. To elucidate the origin of the O radicals in RCH2OH and RCOONa, we employed first-principles calculations to simulate the local DOS of oxygen and the local work function along the a-axis (fractional coordination). The local work function is defined as the local electronic potential minus the Fermi energy of the system. Figures 6(c) and 6(d) show the simulated results for C1–O–C4, C6–OO–Na, and C5–O–C1 indicating that the local DOS of oxygen and the local work functions for C1–O–C4 and C5–O–C1 in RCH2OH are lower and higher, respectively, than those for C6–OO–Na. These results suggest that electrons concentrate on the C1–O–C4 bond as radial C–O electrons in RCH2OH, while the C=O bond in RCOONa induces relaxation of the electron-induced effect. Because the unpaired electrons around the C6 bonds exhibit high reactivity and are rapidly deactivated by secondary reactions,35 they cannot become itinerant electrons. A similar relaxation effect by C=O groups is expected for N–H radicals in the chitin structures of marine products. Thus, materials with high electrical conductivity are believed to become semiconductors, whereas those with low electrical conductivity function as electrical storage bodies. The presence of C=O groups in the structure of a biomaterial is the key factor determining whether it exhibits semiconductor or power storage properties.

FIG. 6.

Molecular structures of (a) RCH2OH and (b) RCOONa with C=O groups as functional groups. (c) Local DOS of oxygen and (d) local work function at positions along the a-axis (fractional coordinates) for C1–O–C4, C6–OO–Na, and C5–O–C1.

FIG. 6.

Molecular structures of (a) RCH2OH and (b) RCOONa with C=O groups as functional groups. (c) Local DOS of oxygen and (d) local work function at positions along the a-axis (fractional coordinates) for C1–O–C4, C6–OO–Na, and C5–O–C1.

Close modal

In this study, we note the electrically polarized water molecules interacting with CNF molecules at the electrode interface. The oxygen atoms in the hydroxyl groups of the CNFs are equatorially arranged in the chair conformation of the glucopyranose rings, forming robust hydrogen bonds with the hydrogen atoms of water molecules when a small amount of water is introduced at the interface between the CNF and its electrode. Notably, Mashl et al.36 reported that water within carbon nanotubes (diameter <0.86 nm) spontaneously crystallizes and transitions into a semiconductor phase, thus demonstrating the semiconducting properties in confined water. At the solid–water interface of confined water, an adsorption layer forms, causing a sudden increase in the mass density and self-diffusion coefficient of water.37 However, in ice, the positions of the oxygen atoms within the water molecule crystal structure are fixed, whereas the positions of the protons are disordered and mobile. Consequently, polarization occurs owing to the orientation of protons in response to an applied electric field.30 By analogy, this suggests that hydrated water at the electrode interface transitions into a pseudo-solid phase within the narrow confinement of 0.311 nm,34 similar to that of an electrolyte. The lone pair electrons of oxygen38 and protons from hydrogen atoms in the pseudo-solid water molecules create an electric double layer with positive charges at the electrode interface.13 This suggests the potential for electrical polarization in the pseudo-solid water molecules at the electrode interface, as shown in Fig. 5(d).

As shown in Fig. 4, the Nyquist diagrams for conifer and RCOONa exhibit markedly different curves: a semicircle for conifer and a trajectory parallel to the imaginary axis (blocking electrode behavior) for RCOONa in the lower frequency region. These results suggest that the semiconductor and normal conductor materials can be represented by electrically lumped [Fig. 5(a)] and distributed [Fig. 5(c)] constant circuits, respectively. In other words, the capacitance at the interface between the cellulose and the electrode in the case of conifer was significantly larger than that in the case of RCOONa. Unlike the large capacitance observed between the electrode and cellulose in conifer, the storage mechanism of RCOONa relies on the nanometer-sized uneven interface of cellulose and its surface texture with a solid electrolyte. Previous studies have attributed the superior power storage effect of RCOONa to its strong electron adsorption effect,9 large power storage capacity,10 and suitable discharge effect.11 In contrast, cellulose and chitosan, which lack C=O functional groups, behave as n-type semiconductors. The junction between these n-type semiconducting materials and electrode materials with large work functions forms a Schottky diode and exhibits a rectifying effect [Figs. 2(a)2(c) and (f)].

Based on the IV characteristics and ESR measurements, the six types of biomaterials can be broadly classified into semiconductors that exhibit rectifying effects and ohmic conductors that do not. The semiconductors lack C=O groups on the side chains of their molecular structures, whereas the ohmic conductors contain these groups. The presence of C=O groups may result in the relaxation of electron-induced effects in cellulose with C1–O–C4 groups and chitosan with N–H groups. AC impedance analysis indicated that the semiconductor and normal conductor materials can be characterized by electrically lumped and distributed constant circuits, respectively.

This work was partially supported by the research project “Cellulose Nanofiber Semiconducting Materials” under the Feasibility Study Program of the New Energy and Industrial Technology Development Organization (NEDO, 23200255-0). We would like to thank Editage (www.editage.jp/) for English language editing.

This work was partially supported by the research project “Cellulose Nanofiber Semiconducting Materials” under the Feasibility Study Program of the New Energy and Industrial Technology Development Organization (NEDO).

M.F. performed the semiconductor analysis and wrote the manuscript. T.Y. and T.S. prepared the CNF sheets and performed the electronic measurements. C.S. performed the ESR measurements. T.H. edited the paper. All the authors discussed the results and commented on the manuscript. M.F. supervised all work.

Mikio Fukuhara: Investigation (lead); Project administration (equal); Writing – original draft (lead). Tomonori Yokotsuka: Formal analysis (equal). Tetsuo Samoto: Investigation (equal). Chika Saito: Formal analysis (equal). Nobuhisa Fujima: Software (lead). Toshiyuki Hashida: Writing – review & editing (lead).

The data that support the findings of this study are available in this article. Additional data are available from the corresponding author on request.

1
A.
Kafy
,
H. C.
Kim
,
L.
Zhai
,
J. W.
Kim
,
L. V.
Hai
,
T. J.
Kang
, and
J.
Kim
,
Sci. Rep.
7
(
1
),
17683
(
2017
).
2
T.
Abitbol
,
A.
Rivkin
,
Y.
Cao
,
Y.
Nevo
,
E.
Abraham
,
T.
Ben-Shalom
,
S.
Lapidot
, and
O.
Shoseyov
,
Curr. Opin. Biotechnol.
39
,
76
88
(
2016
).
3
M.
Abdel-karim
,
A. H.
Salama
, and
M. L.
Hassan
,
J. Phys. Org. Chem.
31
,
e3851
(
2018
).
4
F. J.
Martin-Martinez
,
Proc. Natl. Acad. Sci. U. S. A.
115
(
28
),
7174
7175
(
2018
).
5
H.
Seo
,
H.
Matsumoto
,
S.
Hara
,
M.
Minagawa
,
A.
Tanioka
,
H.
Yako
,
Y.
Yamagata
, and
K.
Inoue
,
Polym. J.
37
(
6
),
391
398
(
2005
).
6
R. A. A.
Muzzarelli
, in
Chitin and Chitosan Hydrogels: Handbook of Hydrocolloids
, edited by
G. O.
Phillips
and
P. A.
Williams
(
Woodhead Publishing Ltd.
,
Cambridge, UK
,
2009
), pp.
849
888
.
7
I. A.
Sogias
,
A. C.
Williams
, and
V. V.
Khutoryanskiy
,
Biomacromolecules
9/7
,
1837
1842
(
2008
).
8
S. K.
Yong
,
M.
Shrivastava
,
P.
Srivastava
,
A.
Kunhikrishnan
, and
N.
Bolan
,
Rev. Environ. Contam. Toxicol.
233
,
1
43
(
2015
).
9
M.
Fukuhara
,
T.
Kuroda
,
F.
Hasegawa
,
T.
Hashida
,
M.
Takeda
,
N.
Fujima
,
M.
Morita
, and
T.
Nakatani
,
Sci. Rep.
11
(
1
),
6436
(
2021
).
10
M.
Fukuhara
,
T.
Yokotsuka
,
T.
Hashida
,
T.
Miwa
,
N.
Fujima
,
M.
Morita
,
T.
Nakatani
, and
F.
Nonomura
,
Sci. Rep.
12
(
1
),
5619
(
2022
).
11
M.
Fukuhara
,
T.
Yokotsuka
,
T.
Takashina
,
N.
Fujima
,
M.
Morita
,
T.
Ito
,
T.
Nakatani
, and
T.
Hashida
,
Sci. Rep.
13
(
1
),
16600
(
2023
).
12
M.
Fukuhara
,
T.
Yokotsuka
,
T.
Hashida
,
F.
Ogawa
,
T.
Sakamoto
,
M.
Takeda
, and
S.
Arai
,
Sci. Rep.
12
(
1
),
21899
(
2022
).
13
M.
Fukuhara
,
T.
Yokotsuka
,
S.
Kayamori
,
A.
Isogai
,
T.
Hashida
, and
A. I. P.
Adv
,
AIP Adv.
14
,
035103
(
2024
).
14
M.
Fukuhara
,
T.
Yokotsuka
,
T.
Samoto
,
M.
Kumadaki
,
M.
Takeda
, and
T.
Hashida
,
Sci. Rep.
14
(
1
),
8692
(
2024
).
15
J. E.
Wertz
and
J. R.
Bolton
,
Electron Spin Resonance: Elementary Theory and Practical Application
(
Chapman & Hall
,
New York
,
1986
).
16
V.
Chechik
,
E.
Carter
, and
D.
Murphy
,
Electron Paramagnetic Resonance
(
Oxford University Press
,
Oxford, UK
,
2016
) https://www.worldcat.org/oclc/945390515.
17
C.
von Sonntag
, “
Free-radical reactions of carbohydrates as studied by radiation techniques
,” in
Advances in Carbohydrate Chemistry and Biochemistry
, edited by
R. S.
Tipson
and
D.
Horton
(
Academic Press
,
New York
,
1980
), Vol.
37
, pp.
7
77
.
18
H. M.
Swartz
,
J. R.
Bolton
, and
D. C.
Borg
,
Biological Applications of Electron Spin Resonance
(
Wiley Interscience
,
New York
,
1972
).
19
J.
Kubo
,
M.
Tsujimura
,
T.
Nakatsubo
, and
H.
Tajima
, “
Preparation of nanofibers from cellulose xanthate
,” in
85th Meeting of Japan Techical Association of Pulp and Paper Industry, 20th of June, 2018
(
TAPPI
,
Japan
,
2018
), pp.
19
32
.
20
T.
Saito
,
S.
Kimura
,
Y.
Nishiyama
, and
A.
Isogai
,
Biomacromolecules
8/8
,
2485
2491
(
2007
).
21
G.
Gutiérrez
and
B.
Johansson
,
Phys. Rev. B
65
(
10
),
4202
(
2002
).
22
S. V.
Patil
and
D. S.
Argyropoulos
,
ChemSusChem
10
(
17
),
3284
3303
(
2017
).
23
J.
Kajihara
,
M.
Enomoto
,
K.
Katoh
,
K.
Mitsuta
, and
M.
Kohno
,
J. Biochem.
104
(
5
),
855
857
(
1988
).
24
A.
Wadsworth
,
H.
Chen
et al,
J. Am. Chem. Soc.
142
(
2
),
652
664
(
2020
).
25
M.
Sathiya
,
J. B.
Leriche
,
E.
Salager
,
D.
Gourier
,
J. M.
Tarascon
, and
H.
Vezin
,
Nat. Commun.
6
,
6276
(
2015
).
26
H.
Kameya
,
S.
Takatsuki
,
R.
Matsuda
,
T.
Tsutsumi
, and
S.
Todoriki
,
J. Food Hyg. Soc. Jpn.
55
,
193
204
(
2014
).
27
W.
Pasanphan
,
G. R.
Buettner
, and
S.
Chirachanchai
,
Carbohydr. Res.
345
(
1
),
132
140
(
2010
).
28
M.
Shimada
,
Y.
Nakanura
,
Y.
Kusama
,
O.
Matsuda
,
H.
Tamura
, and
E.
Kageyama
,
J. Soc. Fiber Sci. Technol., Jpn.
,
30
,
T-299
T-304
(
1974
).
29
S.
Saiki
,
N.
Nagasawa
,
A.
Hiroki
,
N.
Morishita
,
M.
Tamada
,
Y.
Muroya
,
H.
Kudo
, and
Y.
Katsumura
,
Radiat. Phys. Chem.
79
,
276
278
(
2010
).
30
U.
Gryczka
,
D.
Dondi
,
A. G.
Chielewski
,
W.
Migdol
,
A.
Buttafava
, and
A.
Foucitano
,
Radiat. Phys. Chem.
78
,
543
548
(
2009
).
31
T.
Muto
,
M.
Sugihara
,
T.
Goto
, and
Y.
Machi
,
Electronic Materials Devices
(
Ohom
,
Tokyo
,
1986
), p.
161
.
32
K.
Sasaki
and
N.
Shinyashiki
,
Jpn. Soc. Calorim. Therm. Anal.
45
,
124
128
(
2018
).
33
M.
Fukuhara
,
T.
Yokotsuka
,
M.
Morita
,
T.
Ito
,
M.
Yada
,
T.
Nakatani
, and
T.
Hashida
,
Sci. Rep.
14
(
1
),
10350
(
2024
).
34
S. J.
Perkins
,
Eur. J. Biochem.
157
(
1
),
169
180
(
1986
).
35
Y.
Yamauchi
,
M.
Sugio
, and
M.
Kuzuya
,
Chem. Pharm. Bull.
47
,
273
278
(
1999
).
36
R. J.
Mashl
,
S.
Joseph
,
N. R.
Aluru
, and
E.
Jakobsson
,
Nano Lett.
3
,
589
592
(
2003
).
37
Y.
Naruke
,
S.
Kosaka
,
T.
Nakano
,
G.
Kikugawa
, and
T.
Ohara
,
Int. J. Heat Mass Transfer
84
,
584
591
(
2015
).
38
M. A.
Fox
and
J. K.
Whitesell
,
Organic Chemistry
(
Jones and Bartlett Publishers
,
2004
).