How Does the Hydrogen Bonding Interaction Influence the Properties of Polybenzoxazine? An Experimental Study Combined with Computer Simulation

Article

Cite This: Macromolecules XXXX, XXX, XXX −XXX

How Does the Hydrogen Bonding Interaction Inﬂuence the

Properties of Polybenzoxazine? An Experimental Study Combined

with Computer Simulation

Xiaobin Shen,^†,‡Lijun Cao,^†,‡,#Yuan Liu,^†,‡Jinyue Dai,^†,‡Xiaoqing Liu,*^,†,§Jin Zhu,^†,§

and Shiyu Du^†

^†Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, Ningbo, Zhejiang 315201, P. R. China

^‡University of Chinese Academy of Sciences, Beijing 100049, P. R. China

^§Key Laboratory of Bio-based Polymeric Materials Technology and Application of Zhejiang Province, Ningbo, Zhejiang 315201, P.

R. China

S

* Supporting Information

ABSTRACT: The formation of intra- or intermolecular hydrogen bonding and their inﬂuences on polybenzoxazine’s

properties were investigated after the model benzoxazine monomers (o-AB-fbz, o-AF-fbz, o-AP-fbz, p-AB-fbz, p-AF-fbz, p-AP-

fbz) were synthesized. Because of the diﬀerent electron-donating abilities of bridging units (benzene, furan and pyridine) in

isophthalic acid (IPA), 2,5-furandicarboxylic acid (2,5-FDCA) and 2,6-pyridinedicarboxylic acid (2,6-PDCA), the strength of

intramolecular H-bonding involved with the oxygen atom in oxazine ring in o-AB-fbz, o-AF-fbz and o-AP-fbz followed the order

of o-AB-fbz > o-AF-fbz > o-AP-fbz, and the strength of the overall H-bonding was arranged as follows: o-AB-fbz < o-AF-fbz < o-

AP-fbz. While more intermolecular H-bonding was formed in p-AB-fbz and p-AF-fbz as well as p-AP-fbz. DSC and FT-IR results

discovered the relationship between the H-bonding involved with the oxygen atom in oxazine ring and the curing activities of

benzoxazines. After curing reaction, the cured systems showed varied glass transition temperature (T_g), and the inﬂuence of H-

bonding on T_gwas revealed by in situ FT-IR analysis. Molecular dynamics (MD) simulation was also applied to investigate the

properties of synthesized polybenzoxazines and similar results were obtained. Not only the formation of H-bonding but also

their eﬀects on both the curing behaviors of benzoxazine monomers and the thermal properties of cured resins were

systematically investigated, which would help us understand polybenzoxazines more deeply and might be a guideline for

improving their comprehensive properties only by manipulating the H-bonding.

from both industry and academic communities in recent years,

leading to wide applications in the ﬁeld of electronic packaging,

aerospace and transportation et al.^5,6

However, the properties of polybenzoxazines, especially their

high brittleness, low cross-link density, and high curing

temperature, are still subjects to be improved in order to

meet the varied application requirements. Over the past two

decades, numerous eﬀorts have been made to develop the new

INTRODUCTION

■

Polybenzoxazine is a relatively new phenolic-type thermoset-

ting resin that possess a set of desirable properties, including

excellent mechanical and electrical properties, low moisture

absorption, high chemical and thermal resistance and low

ﬂammability.^1,2In addition, its precursor, benzoxazine can be

synthesized from various phenols, amines and formaldehyde

via the Mannich condensation reaction, which ensures its rich

molecular design ﬂexibility.³As for the curing reaction, no

hardeners or catalysts is needed and nearly no byproducts or

toxic volatiles are generated.⁴All of these advantages have

made polybenzoxazine to attract more and more attention

Received: April 6, 2018

Revised: June 18, 2018

A

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Scheme 1. Chemical Structures of Model Benzoxazine Monomers Containing Amide Groups at ortho- and para-Positions with

Respect to the Oxygen Atom (o-AB-fbz, o-AF-fbz, o-AP-fbz, p-AB-fbz, p-AF-fbz, and p-AP-fbz)

polybenzoxazine systems with enhanced or satisﬁed proper-

ties.⁷⁻²⁵And the general strategies can be cataloged into (1)

blending or alloying benzoxazines with other polymers,⁷⁻¹¹(2)

reinforcing polybenzoxazines with ﬁbers¹²⁻¹⁵or inorganic

particles,¹⁶⁻¹⁸(3) functionalizing benzoxazine monomers with

additional reactive groups,¹⁹⁻²³and (4) synthesizing main

chain or side chain polybenzoxazine precursors.^24,25On the

basis of these works, it could summarize that the features of

polybenzoxazines are mainly attributed to the chemical

structures of benzoxazine monomers and the formation of

Mannich based bridges as well as the inter- and intramolecular

H-bonding in the networks after curing reaction.^26,27

determining not only the polymerization temperature and rate

of benzoxazine monomers, but also the thermal and

mechanical properties as well as other unique features of

cured resins. Therefore, the systematical experimental or

theoretical insight into the formation of H-bonding and its

eﬀect on the comprehensive properties of cured resins is a very

essential for polybenzoxazine research.

However, due to the complicated polymerization mecha-

nism of benzoxazines and the complex chemical structures of

cured resins,^28,31,37,38the investigation on the formation or

behaviors of H-bonding in benzoxazine is far from enough,

especially for the cured systems. Sometimes, the hypothesis or

reasoning was more than experimental evidence. As mentioned

above, Ishida et al.^27,29,30conducted some pioneer works on

the scientiﬁc evidence of H-bonding formation in benzoxazine

monomers. With the help of FT-IR³⁹⁻⁴²and NMR,⁴³⁻⁴⁵the

formation of intra- and intermolecular H-bonding in

benzoxazine monomers was identiﬁed and their inﬂuence on

the polymerization initiation were reported. However, because

of the experimental limitation, the strength of H-bonding

interaction and its relationship to the curing reactivity of

benzoxazines were not investigated yet. And what we should

emphasize was that, the experimental evidence about the

relationship between H-bonding strength and thermal proper-

ties of polybenzoxazine networks was rarely reported in

literatures.

Hence, in order to obtain deeper and more fundamental

understanding about the formation and strength of H-bonding

in polybenzoxazine, the model benzoxazine monomers,

containing amide groups at ortho- or para-position with

respect to the oxygen atom in oxazine ring, were carefully

designed and synthesized from isophthalic acid (IPA), 2,5-

furandicarboxylic acid (2,5-FDCA) and 2,6-pyridinedicarbox-

ylic acid (2,6-PDCA) (Scheme 1). It was easy to notice that

the bridging units (benzene, furan and pyridine) in the

benzoxazines possessed diﬀerent electron-donating abilities

(diﬀerent basicity), which could tailor or manipulate the

strength of H-bonding interaction between the amide group

(−CONH−) with other electron-donating groups, such as

−ArOCH₂− and Ar−OH. On the basis of which, not only the

formation of H-bonding, but also their strength could be

characterized and qualitatively compared. Therefore, the eﬀect

Recently, Ishida and his co-workers²⁸⁻³⁰investigated the

catalytic eﬀect of H-bonding involved with the oxygen atom in

oxazine ring and the o-methylol/amide groups. They found

that the H-bonding could accelerate the initiation of

polymerization and slow down the chain propagation.

Especially, the intramolecular ﬁve-membered H-bonding

between o-amide group (−CONH−) and oxazine ring

(−ArOCH₂−) could act like a self-complementary initiator,

which reduced the curing temperature of o-amide benzoxazine

by 60 °C.²⁸The intramolecular interaction between aromatic

hydrogen and oxazine ring, which was responsible for the

curing temperature diﬀerences between ortho-, meta- and para-

positioned bis(benzoxazine)s, was also reported in Endo’s

work.³¹Moreover, the H-bonding interaction was also

employed for the property enhancement of polybenzoxa-

zines.³²⁻³⁶For example, Yagci et al.^32,33prepared several kinds

of self-healing polybenzoxazines taking advantage of inter- or

intramolecular hydrogen bonding in the cured network.

Qutubuddin and his co-workers³⁴prepared chitosan/poly-

benzoxazine cross-linked ﬁlms with improved thermal and

mechanical properties, and the H-bonding might play a crucial

role. Most recently, the higher glass transition temperature

(T_g) of furan-based polybenzoxazine was attributed the more

H-bonding interaction between the oxygen atom in furan ring

and Ar-OH, when compared with the polybenzoxazine derived

from diaminodiphenylmethane (DDM).³⁵And the increased

surface free energy of main-chain type polybenzoxazine

synthesized from diﬀerent aromatic diamines was also

attributed to the increased intermolecular H-bonding in Lin’s

work.³⁶Undoubtedly, H-bonding plays a signiﬁcant role in

B

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

¹H NMR (400 MHz, DMSO-d₆, δ, ppm): 9.87 (s, 2H), 9.73 (s,

2H), 7.69−7.53 (m, 2H), 7.38 (s, 2H), 7.13(m, 2H), 6.96 (d, J = 7.9

Hz, 2H), 6.85 (t, J = 7.3 Hz, 2H).

of H-bonding strength on the thermal properties of

polybenzoxazine could be experimentally investigated. We

believe that the information provided in this paper would help

us understand polybenzoxazine more deeply.

¹³C NMR (400 MHz, DMSO-d₆, δ, ppm): 155.87 (s), 150.20 (s),

148.50 (s), 126.59 (s), 125.35 (s), 124.53 (s), 119.21 (s), 116.04 (s).

N1,N3-Bis(2-hydroxyphenyl)isophthalamide (Compound 2).

Compound 2 was obtained through a method similar to that used

for compound 1, except that 1,3-BDCC (5.89 g, 29.0 mmol) was

taken to replace 2,5-FDCC (yield: 88%).

Furthermore, molecular dynamics (MD) simulation is

becoming increasingly important for the study of polymeric

materials, as it provides a method to numerically conﬁrm the

theoretical models, which help us intuitively observe the

structures of polymers in “molecular size”. As we know, the

early models for cross-linked networks have been created

through static algorithms⁴⁶and stepwise algorithms⁴⁷⁻⁵¹

during the cross-linking reaction. And most recently, MD

simulations have been already tried to investigate the

microstructural properties of polybenzoxazines.^52,53In this

work, the MD simulation was also employed to study the

formation of H-bonding in the model polybenzoaxzines and

predict their properties, which was an useful tool to verify the

experimentally observed properties.

Anal. Calcd for C₂₀H₁₆N₂O₄: C, 68.96; H, 4.63; O, 18.37; N, 8.04.

Found: C, 69.09; H, 4.50; O, 18.47; N, 7.97.

¹H NMR (400 MHz, DMSO-d₆, δ, ppm): 9.74 (d, J = 17.8 Hz,

4H), 8.58 (s, 1H), 8.17 (d, J = 7.6 Hz, 2H), 7.69 (t, J = 6.6 Hz, 3H),

7.07 (t, J = 7.6 Hz, 2H), 6.99−6.90 (m, 2H), 6.86 (t, J = 7.5 Hz, 2H).

¹³C NMR (400 MHz, DMSO-d₆, δ, ppm): 165.06 (s), 149.91 (s),

134.93 (s), 130.86 (s), 129.02 (s), 127.00 (s), 126.20 (s), 125.86 (s),

124.89 (s), 119.26 (s), 116.27 (s).

N2,N6-Bis(2-hydroxyphenyl)pyridine-2,6-dicarboxamide (Com-

pound 3). Compound 3 was synthesized through a method similar

to that used for compound 1, except that 2,6-PDCC (5.75 g, 28.2

mmol) was taken to replace 2,5-FDCC (yield: 85%).

Anal. Calcd for C₁₉H₁₅N₃O₄: C, 65.32; H, 4.33; O, 18.32; N, 12.03.

Found: C, 65.23; H, 4.24; O, 18.43; N, 12.11.

EXPERIMENTAL SECTION

■

Materials. Chemicals and solvents used in this work were

purchased from Aladdin Reagent, China, including isophthalic acid

(IPA) (99%), 2,6-pyridinedicarboxylic acid (2,6-PDCA) (99%), o-

aminophenol (98%), p-aminophenol (98%), trimethylamine (99%),

furfurylamine (FA) (99%), formaldehyde solution (37 wt % in H₂O),

sulfoxide chloride (SOCl₂) (99%), dimethylformamide (DMF)

(99.5%), dimethylacetamide (DMAc) (99%), 1,4-dioxane (99%),

sodium hydroxide (NaOH) (98%) and chloroform (99%), except 2,5-

furandicarboxylic acid (2,5-FDCA) (98%) that was purchased from

Chem target Technologies Co., Ltd. (Mianyang China). All the

chemicals were used as received.

¹H NMR (400 MHz, DMSO-d₆, δ, ppm): 10.44 (s, 2H), 10.06 (s,

2H), 8.43−8.38 (m, 2H), 8.33 (dd, J = 8.6, 6.7 Hz, 1H), 8.05 (dd, J =

7.9, 1.3 Hz, 2H), 7.10−7.01 (m, 2H), 6.99 (dd, J = 7.9, 1.0 Hz, 2H),

6.92−6.86 (m, 2H).

¹³C NMR (400 MHz, DMSO-d₆, δ, ppm): 161.49 (s), 149.33 (s),

148.99 (s), 140.75 (s), 125.90 (d, J = 6.1 Hz), 125.48 (s), 122.60 (s),

119.61 (s), 115.84 (s).

N2,N5-Bis(4-hydroxyphenyl)furan-2,5-dicarboxamide (Com-

pound 4). Compound 4 was obtained by 2,5-FDCC (5.23 g, 27.1

mmol) reacted with p-aminophenol (8.86 g, 81.3 mmol) companied

with triethylamine (15.1 mL, 108.4 mmol), following a method

similar to that used for compound 1 (yield: 89%).

Synthesis of Monomers. 2,5-Furandicarbonyl Dichloride (2,5-

FDCC). 2,5-FDCC was obtained following the method described in

the previous literature.⁵⁴A mixture containing 11.44 g of 2,5-FDCA

(73.4 mmol), 24 mL of SOCl₂(338 mmol), and 200 μL of

dimethylformamide (DMF) was reﬂuxed at 80 °C for 4 h with

constant stirring. The exhaust should be collected with a concentrated

sodium hydroxide aqueous solution through a glass tube from

condenser to a washing bottle, packed with activated silica gel. After

the excess of SOCl₂and DMF was removed under vacuum at room

temperature, the product was isolated and puriﬁed by vacuum

sublimation to get the white crystal (yield: 76%).

Anal. Calcd for C₁₈H₁₄N₂O₅: C, 63.90; H, 4.17; O, 23.65; N, 8.28.

Found: C, 64.07; H, 4.28; O, 23.63; N, 8.08.

¹H NMR (400 MHz, DMSO-d₆, δ, ppm): 10.11 (s, 2H), 9.44 (s,

2H), 7.50 (t, J = 13.1 Hz, 4H), 7.35 (d, J = 10.0 Hz, 2H), 6.80 (t, J =

9.0 Hz, 4H).

¹³C NMR (400 MHz, DMSO-d₆, δ, ppm): 155.26 (s), 154.40 (s),

148.32 (s), 129.19 (s), 122.86 (s), 115.84 (s), 115.38 (s).

N1,N3-Bis(4-hydroxyphenyl)isophthalamide (Compound 5).

Compound 5 was synthesized from 1,3-BDCC (5.96 g, 29.4 mmol)

according to a method similar to that used for compound 4 (yield:

91%).

¹H NMR (400 MHz, CDCl₃, δ, ppm): 7.54 (s, 2H).

1,3-Benzenedicarbonyl Dichloride (1,3-BDCC). 1,3-BDCC was

obtained with the similar method to 2,5-FDCC. 12.33 g of IPA (74.2

mmol) was used for 1,3-BDCC. Then white crystal was obtained after

puriﬁcation (yield: 81%).

Anal. Calcd for C₂₀H₁₆N₂O₄: C, 68.96; H, 4.63; O, 18.37; N, 8.04.

Found: C, 69.13; H, 4.51; O, 18.43; N, 8.01.

¹H NMR (400 MHz, DMSO-d₆, δ, ppm): 10.21 (s, 2H), 9.30 (s,

2H), 8.48 (s, 2H), 8.15−8.03 (m, 2H), 7.72−7.60 (m, 2H), 7.56 (d, J

= 8.8 Hz, 4H), 6.77 (d, J = 8.8 Hz, 4H).

¹H NMR (400 MHz, DMSO-d₆, δ, ppm): 8.47 (s, 1H), 8.14 (dd, J

= 7.8, 1.4 Hz, 2H), 7.60 (t, J = 7.7 Hz, 1H).

¹³C NMR (400 MHz, DMSO-d₆, δ, ppm): 164.53 (s), 153.83 (s),

135.37 (s), 130.60 (s), 130.24 (s), 128.48 (s), 126.75 (s), 122.25 (s),

115.05 (s).

2,6-Pyridinedicarbonyl dichloride (2,6-PDCC). 2,6-PDCC was

obtained with the similar method to 2,5-FDCC. 13.58 g of 2,6-

PDCA (81.3 mmol) was used for 2,6-PDCC. Then white crystal was

obtained after puriﬁcation (yield: 70%).

N2,N6-Bis(4-hydroxyphenyl)pyridine-2,6-dicarboxamide (Com-

pound 6). Compound 6 was synthesized from 2,6-PDCC (5.63 g,

27.6 mmol) according to a method similar to that used for compound

4 (yield: 87%).

¹H NMR (400 MHz, DMSO-d₆, δ, ppm): 8.30 (m, 2H), 8.24 (t, J

= 7.6 Hz, 1H).

N2,N5-Bis(2-hydroxyphenyl)furan-2,5-dicarboxamide (Com-

pound 1). Compound 1 was synthesized by reacting o-aminophenol

with 2,5-FDCC and triethylamine in DMAc.⁵⁵o-Aminophenol (9.13

g, 83.6 mmol) was mixed with triethylamine (15.5 mL, 111.6 mmol)

in 100 mL of DMAc. Then 2,5-FDCC (5.38 g, 27.9 mmol) dissolved

in 50 mL of DMAc was added drop by drop into the mixture. After

that the solution was cooled down to 0 °C and kept at this

temperature for 4 h, the reaction mixture was poured into cold water

and the yellow precipitate was collected via ﬁltration. Finally, it was

heated at 50 °C for 8 h in the vacuum oven and the target product

was obtained (yield: 84%).

Anal. Calcd for C₁₉H₁₅N₃O₄: C, 65.32; H, 4.33; O, 18.32; N, 12.03.

Found: C, 65.19; H, 4.34; O, 18.40; N, 12.15.

¹H NMR (400 MHz, DMSO-d₆, δ, ppm): 10.85 (s, 2H), 9.40 (s,

2H), 8.34 (t, J = 14.1 Hz, 2H), 8.26 (dd, J = 8.4, 7.0 Hz, 1H), 7.64 (d,

J = 8.8 Hz, 4H), 6.83 (d, J = 8.8 Hz, 4H).

¹³C NMR (400 MHz, DMSO-d₆, δ, ppm): 161.45 (s), 154.58 (s),

149.27 (s), 139.99 (s), 129.66 (s), 125.06 (s), 123.51 (s), 115.37 (s).

N2,N5-Bis(3-(furan-2-ylmethyl)-3,4-dihydro-2H-benzoxazine-8-

yl)furan-2,5-dicarboxamide (o-AF-fbz). A 3.40 g sample of furfuryl-

amine (35.1 mmol) was dissolved in 250 mL of 1,4-dioxane before it

was cooled to 0 °C. Then 5.9 mL of formaldehyde (80.1 mmol) was

added dropwisely and the reaction was conducted at 0 °C for one

Anal. Calcd for C₁₈H₁₄N₂O₅: C, 63.90; H, 4.17; O, 23.65; N, 8.28.

Found: C, 64.13; H, 4.21; O, 23.43; N, 8.11.

C

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

more hour. After the as-synthesized compound 1 (5.12 g, 15.1 mmol)

was added, the reaction temperature was increased up to 105 °C and

maintained at this temperature for 20 h. Then the solvent was

evaporated under vacuum. The obtained liquid was washed by sodium

hydroxide (4% w/w) to eliminate the residual starting materials and

dried over anhydrous magnesium sulfate. At last, the precipitate was

collected by ﬁltration and dried in the vacuum oven to obtain the light

yellow powder (yield: 29%).

120.47 (s), 117.21 (s), 116.57 (s), 110.71 (s), 109.53 (s), 82.30 (s),

49.97 (s), 48.66 (s).

N1,N3-Bis(3-(furan-2-ylmethyl)-3,4-dihydro-2H-benzoxazine-6-

yl)isophthalamide (p-AB-fbz). p-AB-fbz was synthesized from

compound 5 (5.16 g, 14.8 mmol) following following a synthesis

procedure similar to that used for p-AF-fbz (yield: 16%).

Anal. Calcd for C₃₄H₃₀N₄O₆: C, 69.14; H, 5.12; O, 16.25; N, 9.49.

Found: C, 69.22; H, 5.14; O, 16.35; N, 9.31.

FT-IR (KBr, cm⁻¹): 928, 1340, and 1363.

Anal. Calcd for C₃₂H₂₈N₄O₇: C, 66.20; H, 4.86; O, 19.29; N, 9.65.

Found: C, 66.19; H, 4.84; O, 19.40; N, 9.55.

¹H NMR (400 MHz, CDCl₃, δ, ppm): 8.34 (s, 1H), 8.15 (s, 2H),

8.00 (d, J = 7.6 Hz, 2H), 7.56 (t, J = 7.7 Hz, 1H), 7.42 (s, 4H), 7.24

(s, 2H), 6.79 (d, J = 8.7 Hz, 2H), 6.39−6.33 (m, 2H), 6.27 (d, J = 3.0

Hz, 2H), 4.87 (s, 4H), 4.00 (s, 4H), 3.91 (s, 4H).

FT-IR (KBr, cm⁻¹): 925, 1340, and 1362.

¹H NMR (400 MHz, CDCl₃, δ, ppm): 8.58 (s, 2H), 8.32 (d, J = 7.5

Hz, 2H), 7.40 (t, J = 4.0 Hz, 2H), 7.31 (d, J = 5.2 Hz, 2H), 6.96 (dd, J

= 18.4, 10.5 Hz, 2H), 6.77 (d, J = 7.0 Hz, 2H), 6.32 (dt, J = 10.5, 5.3

Hz, 2H), 6.27 (t, J = 8.8 Hz, 2H), 5.01 (s, 4H), 4.03 (s, 4H), 3.92 (s,

4H).

¹³C NMR (400 MHz, CDCl₃, δ, ppm): 165.29 (s), 151.90 (s),

151.62 (s), 143.11 (s), 135.81 (s), 131.19 (s), 130.77 (s), 129.76 (s),

125.91 (s), 121.16 (s), 120.60 (s), 117.33 (s), 110.71 (s), 109.55 (s),

82.35 (s), 50.11 (s), 48.68 (s).

¹³C NMR (400 MHz, CDCl₃, δ, ppm): 155.23 (s), 151.29 (s),

149.03 (s), 142.85 (s), 125.70 (s), 123.06 (s), 120.88 (s), 119.41 (s),

118.78 (s), 116.60 (s), 110.30 (s), 109.23 (s), 83.03 (s), 49.11 (s),

48.53 (s).

N2,N6-bis(3-(furan-2-ylmethyl)-3,4-dihydro-2H-benzoxazine-6-

yl)pyridine-2,6-dicarboxamide (p-AP-fbz). p-AP-fbz was synthesized

from compound 6 (5.24 g, 15.0 mmol) following following a synthesis

procedure similar to that used for p-AF-fbz (yield: 17%).

Anal. Calcd for C₃₃H₂₉N₅O₆: C, 67.00; H, 4.94; O, 16.23; N, 11.84.

Found: C, 66.85; H, 5.09; O, 16.46; N, 11.73.

N1,N3-Bis(3-(furan-2-ylmethyl)-3,4-dihydro-2H-benzoxazine-8-

yl)isophthalamide (o-AB-fbz). o-AB-fbz was obtained following a

synthesis procedure similar to that used for o-AF-fbz, except for the

as-synthesized compound 2 (5.29 g, 15.2 mmol) was taken to replace

compound 1. After puriﬁed and dried in vacuum oven, a yellow

powder was obtained (yield: 32%).

FT-IR (KBr, cm⁻¹): 928, 1340, and 1363.

¹H NMR (400 MHz, CDCl₃, δ, ppm): 9.45 (s, 2H), 8.44 (t, J = 9.2

Hz, 2H), 8.10 (t, J = 7.8 Hz, 1H), 7.54 (d, J = 2.3 Hz, 2H), 7.42 (d, J

= 1.1 Hz, 2H), 7.33 (dd, J = 8.7, 2.4 Hz, 2H), 6.81 (t, J = 8.5 Hz, 2H),

6.33 (dd, J = 10.3, 8.4 Hz, 2H), 6.26 (d, J = 3.0 Hz, 2H), 4.88 (s, 4H),

4.02 (s, 4H), 3.92 (s, 4H).

Anal. Calcd for C₃₄H₃₀N₄O₆: C, 69.14; H, 5.12; O, 16.25; N, 9.49.

Found: C, 69.19; H, 5.22; O, 16.21; N, 9.44.

FT-IR (KBr, cm⁻¹): 925, 1340, and 1362.

¹³C NMR (400 MHz, CDCl₃, δ, ppm): 161.23 (s), 151.59 (s),

151.42 (s), 149.23 (s), 142.77 (s), 139.64 (s), 130.32 (s), 125.57 (s),

120.70 (s), 120.24 (s), 117.06 (s), 110.39 (s), 109.19 (s), 82.10 (s),

49.79 (s), 48.41 (s).

¹H NMR (400 MHz, CDCl₃, δ, ppm): 8.51 (s, 2H), 8.45 (s, 1H),

8.38 (d, J = 8.0 Hz, 2H), 8.10 (dd, J = 7.8, 1.7 Hz, 2H), 7.65 (dd, J =

16.0, 8.2 Hz, 1H), 7.41 (t, J = 7.1 Hz, 2H), 6.96 (t, J = 7.9 Hz, 2H),

6.77 (d, J = 7.3 Hz, 2H), 6.34 (dd, J = 3.1, 1.9 Hz, 2H), 6.26 (d, J =

3.0 Hz, 2H), 5.02 (s, 4H), 4.07 (s, 4H), 3.96 (s, 4H).

¹³C NMR (400 MHz, CDCl₃, δ, ppm): 164.35 (s), 151.30 (s),

142.79 (s), 135.86 (s), 130.24 (s), 129.35 (s), 126.33 (s), 122.66 (s),

120.68 (s), 119.22 (s), 118.42 (s), 110.26 (s), 109.12 (s), 82.88 (s),

49.00 (s), 48.38 (s).

Curing Procedure. Benzoxazine monomers were melted and

transferred to a stainless steel mold as quickly as possible. Then the

sample was cured by the programmed temperature rising method: 2 h

at 180 °C, 2 h at 200 °C, and 4 h at 220 °C. After curing reaction, the

samples were cooled down to the room temperature slowly to prevent

cracking. In this work, all of the samples including o-AF-fbz, o-AB-fbz,

o-AP-fbz, p-AF-fbz, p-AB-fbz, and p-AP-fbz were cured under the

same condition.

Measurements. NMR spectra were recorded on a 400 MHz

Bruker AVANCE III spectrometer. It was conducted at 25 °C using

DMSO-d₆or CDCl₃as the solvent. The tetramethysilane (TMS) was

used as internal standard and the times of scans was 16. For the 2D

¹H−¹H NOESY NMR measurement, related parameters were set as

2048 data points along the f 2 dimension, 256 free induction decays in

the f1 dimension, and relaxation delay of 2 s, and the mixing time was

700 ms.

Mass spectra were measured with a LC-Q-TOF (AB Sciex,

America) using ionization at 500 °C with an ionization voltage of

5500 V. Elementary analysis (EA) was measured by Elementar

(Elementar, Germany).

Fourier transform infrared (FT-IR) spectra were recorded in

transmission mode using a Thermo Nicolet 6700 Fourier transform

infrared spectrometer (Thermo-Fisher Scientiﬁc). Samples were

powdered and dispersed into a KBr matrix with a weight

concentration of about 1 wt %. Spectra were scanned from 400 to

4000 cm⁻¹with 32 scans collected for each sample. For the in situ FT-

IR measurement at various temperature (50, 100, 150, 200, and 250

°C). Cured samples were inserted into a hot cell which was adapted

to the FT-IR spectrometer, and the scan was started as soon as the hot

cell temperature reached the desired temperature.

The DSC measurement was conducted on a METTLER

TOLEDO-TGA/DSC I under a nitrogen atmosphere ﬂow rate of

20 mL min⁻¹. Approximately 5−10 mg of each sample was weighed

and sealed in 40 μL aluminum crucibles. The sample was heated from

50 to 300 °C at the heating rate of 10 °C min⁻¹. In order to determine

the activation energy of polymerization, the DSC measurement was

conducted at diﬀerent heating rates of 2,5, 10, and 20 °C min⁻¹. DSC

N2,N6-Bis(3-(furan-2-ylmethyl)-3,4-dihydro-2H-benzoxazine-8-

yl)pyridine-2,6-dicarboxamide (o-AP-fbz). o-AP-fbz was obtained

following a synthesis procedure similar to that used for o-AF-fbz,

except for the as-synthesized compound 3 (5.17 g, 14.8 mmol) that

was taken to replace compound 1. After the reaction product was

puriﬁed and dried under vacuum, a yellow powder was achieved

(yield: 26%).

Anal. Calcd for C₃₃H₂₉N₅O₆: C, 67.00; H, 4.94; O, 16.23; N, 11.84.

Found: C, 66.70; H, 5.02; O, 16.41; N, 11.76.

FT-IR (KBr, cm⁻¹): 925, 1340, and 1362.

¹H NMR (400 MHz, CDCl₃, δ, ppm): 9.97 (s, 2H), 8.49 (d, J = 7.8

Hz, 2H), 8.35 (t, J = 7.5 Hz, 2H), 8.17−8.10 (m, 1H), 7.36−7.30 (m,

2H), 7.02−6.96 (m, 2H), 6.79 (d, J = 7.6 Hz, 2H), 6.28−6.23 (m,

2H), 6.19 (t, J = 6.4 Hz, 2H), 4.95 (s, 4H), 4.01 (s, 4H), 3.90 (s, 4H).

¹³C NMR (400 MHz, CDCl₃, δ, ppm): 161.49 (s), 151.27 (s),

149.63 (s), 143.82 (s), 142.76 (s), 139.48 (s), 125.90 (s), 125.59 (s),

123.22 (s), 120.81 (s), 119.53 (s), 110.35 (s), 109.23 (s), 83.08 (s),

49.21 (s), 48.44 (s).

N2,N5-Bis(3-(furan-2-ylmethyl)-3,4-dihydro-2H-benzoxazine-6-

yl)furan-2,5-dicarboxamide (p-AF-fbz). p-AF-fbz was synthesized

from compound 4 (5.02 g, 14.9 mmol) following a synthesis

procedure similar to that used for o-AF-fbz (yield: 21%).

Anal. Calcd for C₃₂H₂₈N₄O₇: C, 66.20; H, 4.86; O, 19.29; N, 9.65.

Found: C, 66.15; H, 4.91; O, 19.36; N, 9.59.

FT-IR (KBr, cm⁻¹): 928, 1340, and 1363.

¹H NMR (400 MHz, CDCl₃, δ, ppm): 8.80 (s, 2H), 7.44 (s, 2H),

7.41 (d, J = 1.0 Hz, 2H), 7.23 (s, 2H), 6.75 (d, J = 8.7 Hz, 2H), 6.35−

6.32 (m, 2H), 6.25 (d, J = 3.0 Hz, 2H), 4.84 (s, 4H), 3.98 (s, 4H),

3.89 (s, 4H).

¹³C NMR (400 MHz, CDCl₃, δ, ppm): 156.15 (s), 151.86 (s),

151.64 (s), 149.05 (s), 143.06 (s), 130.62 (s), 121.27 (s), 120.79 (s),

D

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Scheme 2. Synthetic Route for the Model Benzoxazine Monomers Containing Amide Groups at ortho- and para-Positions with

Respect to the Oxygen Atom (o-AF-fbz, o-AB-fbz, o-AP-fbz, p-AF-fbz, p-AB-fbz, and p-AP-fbz)

Figure 1. Variation of the chemical shift of NH in o-AB-fbz (a), o-AF-fbz (a), o-AP-fbz (a), p-AB-fbz (b), p-AF-fbz (b), and p-AP-fbz (b) as a

function of concentration in CDCl₃at 25 °C.

analysis was repeated twice for each sample to ensure the accuracy.

TGA measurement was conducted on a Mettler-Toledo TGA/DSC1

RESULTS AND DISCUSSION

■

Synthesis and Characterization of Model Benzox-

thermogravimetric analyzer (METTLER TOLEDO, Switzerland)

with high purity nitrogen or air as purge gas (50 mL min⁻¹) at a

scanning rate of 20 °C min⁻¹from 50 to 800 °C. DMA was

performed on Mettler-Toledo DMAQ800 under a tensile mode at a

frequency of 1 Hz with the amplitude of 5 μm. The test samples were

scanned from 0 to 360 °C at a heating rate of 3 °C min⁻¹. Each

sample was tested three times to ensure the accuracy.

azine Monomers. Scheme 2 illustrated the synthesis of o-AF-

fbz, o-AB-fbz, o-AP-fbz, p-AF-fbz, p-AB-fbz, and p-AP-fbz.

Started from diﬀerent dicarboxylic acids (IPA, 2,5-FDCA and

2,6-PDCA), six kinds of bisphenol monomers containing

amide group were synthesized. Then, they were reacted with

formaldehyde in the solution of 1,4-dioxane and the target

products were obtained. The detailed synthesis procedures

were described in the Experimental Section, and the

Simulation. The MD simulations are carried out with the

Materials Studio 7.0 (Accelrys. Inc.).⁵³The partial charges on the

molecules are obtained from iterative partial equalization of orbital

electronegativity via Gasteiger method.⁵⁴The nonbonded interactions

are calculated from the Lennard-Jones (12−6) potential function and

group-based summation for van der Waals and electrostatic

interactions, respectively, with the cutoﬀ set at 1.2 nm for both

types of interactions. The temperature of simulation box is controlled

1

Supporting Information, Figure S1−S12, showed the H and

¹³C NMR spectra of all the synthesized compounds. Besides

this , the mass spectrometry of the six model benzoxazine

monomers, o-AF-fbz, o-AB-fbz, o-AP-fbz, p-AF-fbz, p-AB-fbz,

and p-AP-fbz were also collected in Supporting Information

(Figure S13−S18)). These chemical structure conﬁrmation

indicated the high purity of ﬁnal products and ensured the

investigation on the relationship between their chemical

structures and curing behaviors.

́

using Nose−Hoover−Langevin (NHL) thermostat with a Q ratio of

0.01.⁵⁵For the pressure control in NPT-MD simulations (constant

pressure, temperature and number of particles), the Parrinello−

Rahman barostat was taken.⁵⁶⁻⁵⁸All the model systems in this study

consist of 100 monomer molecules per simulation box for each

studied resin. The size of the simulation box is ∼5 nm. The cross-link

networks are constructed using the cyclic polymerization atomistic

model with conducted multistep topology relaxation. The detailed

procedures are following the previous work.⁴⁸

On the basis of the chemical structure analysis of o-AF-fbz,

o-AB-fbz, o-AP-fbz, p-AF-fbz, p-AB-fbz, and p-AP-fbz, it was

easy to speculate the presence of inter- or intramolecular H-

bonding in them, easily formed between the electron acceptor

E

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

a

Scheme 3. Proposed H-Bonding Interactions in o-AB-fbz, o-AF-fbz, o-AP-fbz, p-AB-fbz, p-AF-fbz, and p-AP-fbz

a

Green square-dotted intervening with round-dotted lines depict intermolecular H-bonding, while blue round-dotted lines represent the

intramolecular H-bonding. (For easy identiﬁcation, the H-bondings related to the furfurylamine structure as well as other possible H-bondings were

not illustrated, due to their low strength.)

1

of −CONH− and electron donors of O in furan ring and N in

pyridine ring. As we know, for the ¹H NMR spectrum, the used

solvent and concentration as well as temperature will strongly

inﬂuence the signal quality and chemical shift of NH. Also, the

formation of H-bonding is another factor to determine its

chemical shift. For example, in the case of intermolecular H-

bonding, the chemical shift of NH will move to a higher ppm

with the increased concentration because of the decreased

average of hydrogen-bonding distance. While the intra-

molecular hydrogen-bonded systems will not be aﬀected by

the concentration. Therefore, through the H NMR experi-

ments, the inter- and intramolecular H-bonding in the solution

could be analyzed.^27,29,30,41For instance, Ishida and his co-

workers²⁹once characterized the chemical shifts of OH in

methylol groups at the ortho- or para-position with respect to

the oxazine ring as a function of concentration in CDCl₃at 25

°C. On the basis of which, they conﬁrmed that the signals for

intramolecular hydrogen bonded OH showed nearly no change

upon the increment of concentration. While the OH signals

related to the intermolecular H-bonding demonstrated a

F

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

signiﬁcant chemical shift toward downﬁeld when the

concentration was increased. In this work, the variation of

the chemical shift of NH at diﬀerent solution concentration

was also studied. Figure 1a showed the chemical shifts of NH

in o-AB-fbz, o-AF-fbz, and o-AP-fbz as a function of

bonding could be formed between the NH groups and

nitrogen in pyridine as well as oxygen atom in the furan ring.

At higher concentrations, the intermolecular H-bonding would

form when the intermolecular distance became short enough.

Therefore, the intermolecular and intramolecular hydrogen

bondings existed simultaneously. However, in the system of p-

AB-fbz, only intermolecular hydrogen bonding could be

formed.

1

concentration, obtained from H NMR spectra using CDCl₃

as the solvent at 25 °C. The independence of the NH chemical

shift in o-AB-fbz, o-AF-fbz, and o-AP-fbz with respect to

concentration clearly indicated that no intermolecular H-

bonding were formed within the studied concentration range.

Moreover, the high chemical shifts of these particular NH

groups (8.51 ppm for o-AB-fbz, 8.58 ppm for o-AF-fbz and

9.97 ppm for o-AP-fbz) suggested that the protons were, in

fact, participating in the intramolecular H-bonding.^27,30And it

might be due to the higher negative electron density of N atom

in o-AP-fbz, the proton in amide group (−CONH−) was more

strongly deshielded and its chemical shift was high up to 9.97

ppm, which might indicate the stronger H-bonding. In the case

of o-AB-fbz and o-AF-fbz, the interaction between the proton

(−CONH−) and O atoms (no matter it is in the furan ring or

oxazine ring) might be same. Therefore, nearly no diﬀerence in

the chemical shift of NH was observed. Diﬀerent from Figure

1a, Figure 1b indicated that the chemical shift of the NH group

in p-AB-fbz, p-AF-fbz, and p-AP-fbz was inﬂuenced in diﬀerent

degree by the solution concentration. In particualr, in p-AB-fbz,

Although above results conﬁrmed that NH groups in o-AB-

fbz, o-AF-fbz and o-AP-fbz participated in intramolecular H-

bonding, there was no indication about the possibility of

breaking or disrupting the H-bonding interactions. In other

words, no information had yet been obtained to clarify whether

all of the NH protons participated in the formation of H-

bonding or not. This fact was related to the strength or stability

of the intramolecular H-bonding in o-AB-fbz, o-AF-fbz, and o-

1

AP-fbz. To address this question, H−¹H nuclear overhauled

eﬀect spectroscopy (NOESY), a 2D NMR technique which

can provide structural information by through-space inter-

actions,^27,59,60was carried out and the results were presented

in Figure 2.

In o-AB-fbz, besides the H-bonding, the protons in amide

group (−CONH) also have interactions with the proton at

position-1 (H₁) in oxazine ring, aromatic protons at position-2

(H₂) and position-3 (H₃), which was supported by the signals

in Figure 2a (marked as dots 1, 2, 3 in red type, respectively).

Similar interactions were also observed for o-AF-fbz (Figure

2b) and o-AP-fbz (Figure 2c). This result indicated the

presence of free NH protons, which did not participate in the

H-bonding formation. As we know, the signal intensity reﬂects

the strength or intensity of through-space interaction. The

higher signal intensity, the stronger or more interaction.⁴⁸

When the intensities of signal 1 and signal 2 in parts a−c

ofFigure 2 were compared, from o-AB-fbz to o-AP-fbz, the

through-space interaction followed the order of o-AB-fbz > o-

AF-fbz > o-AP-fbz and demonstrated that the strength of

overall intramolecular H-bonding was in the order of o-AB-fbz

< o-AF-fbz < o-AP-fbz. This result was reasonable based on

their chemical structures analysis. In o-AF-fbz and o-AP-fbz,

besides the oxygen atom in oxazine ring, the oxygen atom in

furan ring and nitrogen atom in pyridine ring would also tend

to participate in the ﬁve-membered H-bonding formation with

NH group. The NH proton attraction competition between

these diﬀerent electron-donors was inevitable and due to the

higher basicity of pyridine ring, the RCON−H--N (pyridine)

interaction might be stronger than RCON−H--O (furan).

Therefore, the strength of intramolecular hydrogen bonding

between NH group and oxygen in oxazine ring would be

diﬀerent, which would dramatically inﬂuence the curing

behaviors of benzoxazines.^27,29,30And in the following section,

the curing behaviors of synthesized model benzoxazine

precursors will be investigated. For easy identiﬁcation, the

possible diﬀerent intra- and intermolecular H-bonding as well

as their strength comparison in o-AB-fbz, o-AF-fbz, o-AP-fbz, p-

AB-fbz, p-AF-fbz, and p-AP-fbz were brieﬂy illustrated in

Figure 3. From red to violet, the strength of diﬀerent type H-

bonding was qualitatively represented by the diﬀerent colors.

Curing Behaviors of Model Benzoxazines. The curing

behaviors of o-AB-fbz, o-AF-fbz, o-AP-fbz, p-AB-fbz, p-AF-fbz,

and p-AP-fbz were investigated by DSC. As marked in Figure

4, the peak exothermic temperatures for o-AB-fbz, o-AF-fbz, o-

AP-fbz, p-AB-fbz, p-AF-fbz, and p-AP-fbz were 211, 218, 227,

233, 238, and 241 °C, respectively. And the onset curing

1

the H NMR signal for NH was shifted from 7.8 to 8.7 ppm

when the concentration was increased from 15 to 65 Mm. And

in the system of p-AF-fbz and p-AP-fbz, the signals for NH

showed a relatively weak dependence on the concentration,

indicated by the smaller slopes of red and blue lines in Figure

1b. In Ishida’s work,²⁹they also reported the similar results

that the signal for OH in ortho-positioned methylol groups was

independent of the concentration at ﬁrst and when the

concentration was higher than 2.5 mM, the weak dependence

was noticed. And this result was attributed to the competition

between inter- and intramolecular H-bonding. That was to say

the inter- and intramolecular hydrogen bonding existed

simultaneously. In our work, the dependence of NH chemical

shift for p-AB-fbz, p-AF-fbz, and p-AP-fbz upon concentration

was undoubtedly an indication for the intermolecular H-

bonding. Besides that, the varying degrees of concentration

dependence for NH chemical shift indicated the simultaneous

existence of inter- and intramolecular H-bonding as well as the

diﬀerent ratio of them in p-AB-fbz, p-AF-fbz, and p-AP-fbz.

The fraction of intermolecular H-bonding should follow the

order of p-AB-fbz > p-AF-fbz > p-AP-fbz, based on the degree

of concentration dependence showing in Figure 1b.

Scheme 3 depicted the proposed inter- and intramolecular

H-bonding interactions mainly involved with the various

central moieties (benzene, furan and pyridine moieties) and

amide group in o-AB-fbz, o-AF-fbz, o-AP-fbz, p-AB-fbz, p-AF-

fbz and p-AP-fbz. For easy identiﬁcation, the H-bonding

related to furfurylamine structure as well as other possible H-

bonding were not illustrated, due to their low strength. In the

case of o-AB-fbz, o-AF-fbz, and o-AP-fbz, the NH groups

tended to participate in intramolecular H-bonding with the

oxygen atom in oxazine ring, oxygen in furan or nitrogen in

pyridine ring. And due to the formation of strong and stable

ﬁve-membered ring, the intermolecular H-bonding interactions

were diﬃcult to take place. That was to say the intramolecular

H-bonding was dominant in o-AB-fbz, o-AF-fbz and o-AP-fbz,

which was consistent with the results from Figure 1a. As for p-

AP-fbz and p-AF-fbz, the ﬁve-membered intramolecular H-

G

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Figure 3. Illustration of diﬀerent intra- and intermolecular hydrogen

bonding and the qualitative strength comparison between them in o-

AB-fbz, o-AF-fbz, o-AP-fbz, p-AB-fbz, p-AF-fbz, and p-AP-fbz.

(RCON−H--O from oxazine ring as an example for intermolecular

hydrogen bonding).

Figure 4. DSC heating curves of diﬀerent precursors.

Figure 2. 2D ¹H−¹H NOESY NMR spectra of o-AB-fbz (a), o-AF-fbz

(b), and o-AP-fbz (c) in CDCl₃.

thermograms did not show any melting endotherms, possibly

attributed to the inﬂuence of amide groups,⁵⁵the heating

curves of o-AB-fbz and p-AB-fbz after several times puriﬁcation

showed similar peak exothermic temperatures, which con-

ﬁrmed the high purity of the synthesized monomers again. As

we know, under the same curing condition, the onset and peak

curing temperature can be taken as an indicator for curing

reactivity evaluation. The higher the onset or peak curing

temperature, the lower curing reactivity.⁶¹Apparently, the

ortho-positioned systems (o-AB-fbz, o-AF-fbz, and o-AP-fbz)

showed higher curing reactivity when compared with those of

the para-substituted systems (p-AB-fbz, p-AF-fbz, and p-AP-

fbz). In addition, for the ortho-substituted systems, their curing

activities followed the order of o-AB-fbz > o-AF-fbz > o-AP-fbz.

And the para-substituted systems were arranged as follows: p-

AB-fbz > p-AF-fbz > p-AP-fbz. It was noted that the curing

temperatures of them followed the same order (175 °C for o-

AB-fbz, 183 °C for o-AF-fbz, 188 °C for o-AP-fbz, 207 °C for

p-AB-fbz, 213 °C for p-AF-fbz and 218 °C for p-AP-fbz). What

should be mentioned was that the melting endotherms, usually

taken as a parameter for monomer purity evaluation, were

absent in the DSC scan. It is well-known that, the impurities

will inﬂuence the performance of cured benzoxazine resins

seriously, including the curing property. Besides the NMR, FT-

IR, elementary analysis and mass spectra (Supporting

Information, Figures S13−S18) measurements, in order to

further conﬁrm the purity of the synthesized monomers, o-AB-

fbz and p-AB-fbz were applied for several puriﬁcation

treatments and after each puriﬁcation, the DSC measurement

was conducted and the curves were collected for comparison

(Supporting Information, Figure S19). Although all the DSC

H

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

−1

Figure 5. Plots of −ln(qT ^‑2) versus 1000 R⁻¹T_p(a) and ln q versus 1052 R⁻¹T_p(b).

p

activities followed the same order of H-bonding strength

involved with the oxygen atoms in oxazine ring, rather than the

strength of overall H-bonding, in the model benzoxazines.⁶²

In order to further conﬁrm the curing reactivity comparison,

the Kissinger and Ozawa methods, which are considered to be

the simple approaches to quantify a complex thermoset curing

process, were employed.⁶³⁻⁶⁵For research convenience, only

the para-substituted systems (p-AB-fbz, p-AF-fbz and p-AP-

fbz) were applied for the Kissinger and Ozawa method

investigation. The Kissinger method⁶³is based on a linear

relationship between logarithm of (qT_p⁻²) and T_p⁻¹and can be

explained by the following eq 1:

Table 1. E_aComparison for para-Substituted Benzoxazine

Determined by Kissinger and the Modiﬁed Ozawa Equation

a

E_a(kJ mol⁻¹

)

ln A

heating rates

(q)

T_p

(K) Kissinger Ozawa Kissinger Ozawa

samples

(°C min⁻¹

)

p-AB-fbz

2

5

10

20

2

475

489

499

506

484

492

502

511

488

496

505

514

137.0

162.1

173.6

138.0

162.2

173.9

32.72

38.80

41.13

32.64

38.42

40.63

p-AF-fbz

p-AP-fbz

5

10

20

2

5

10

i

j

y

z

i

j

y

z

q

E_a

RT_p

AR

E_a

ln

= lnj z −

j

z

²z

z

T_p

k

{

(1)

k

{

where T_pwas the peak exothermic temperature (K), q was the

heating rate (°C min⁻¹), E_awas an average activation energy of

the curing reaction (J mol⁻¹), A was the pre-exponential factor

(min⁻¹), and R was the universal gas constant with a value of

8.314 J mol⁻¹K⁻¹. T_pand its corresponding heating rate (q)

could be obtained from Figure S20 (Supporting Information)

and the activation energy could be calculated from the slope of

20

a

The dimension of A is min⁻¹, and α was 0.5.

eration eﬀect of intramolecular ﬁve-membered H-bonding

formed between the NH in amide group and the oxygen in

oxazine ring. In this work, the intra- and intermolecular H-

bonding in the model benzoxazines was clearly identiﬁed and

the higher curing reactivity of ortho-substituted systems was

undoubtedly related to the H-bonding. As illustrated in Figure

3, in the o-amide systems (o-AB-fbz, o-AF-fbz, and o-AP-fbz),

not only the oxygen in the oxazine ring participated in the

intramolecular hydrogen bonding, but also the nitrogen atom

in pyridine and oxygen in furan ring did. And due to the higher

proton withdrawing ability of pyridine, the RCON−H--N

(pyridine) interaction was stronger than that of RCON−H--O

(furan) and RCON−H--O (Ar). Therefore, the strength of the

ﬁve-membered intramolecular H-bonding formed between

RCONH and the oxygen in oxazine ring should follow the

order of o-AB-fbz > o-AF-fbz > o-AP-fbz. On the basis of the

proposed curing mechanism in literatures,^{23,27,29−31}the

catalytic eﬀect of H-bonding in the diﬀerent systems should

follow the same order and the sequence of o-AB-fbz > o-AF-fbz

> o-AP-fbz for their curing activities was reasonable. In Scheme

4, the proposed mechanism was illustrated brieﬂy. In the case

of p-amide systems (p-AB-fbz, p-AF-fbz, p-AP-fbz), due to the

presence of furan and pyridine in p-AF-fbz and p-AP-fbz,

respectively. The intramolecular H-bonding (RCON−H--N

(pyridine) and RCON−H--O (furan)) could be formed, which

would attract and limit the movement of partial protons in

RCON-H group, leading to the decreased possibility of

intermolecular H-bonding between the oxygen in oxazine

the plots of −ln(qT_p⁻²) versus T_p(Figure 5a).

−1

The modiﬁed Ozawa method⁶⁴is another way to determine

the E_a, based on eq 2:

i

y

AE

R

j E_a

z

i

j

y

z

a

j

z

ln q = ln

− ln g(α) − 5.331 − 1.052

j

z

RT_p

k

{

(2)

k

{

α

dα

where g(α) =

= −ln(1 − α) was the integral con-

∫

f(α)

0

version function. E_acould be derived from the slope of ln q

−1

against T_pplots (Figure 5b). The diﬀerent heating rates (q)

and corresponding T_pas well as E_aobtained via Kissinger and

modiﬁed Ozawa methods were summarized in Table 1. It was

noticed that the activation energy of p-AB-fbz was the lowest,

indicating the highest curing reactivity. While p-AP-fbz showed

the lowest curing activity indicating by the highest activation

energy. This was in a good agreement with the result obtained

from Figure 4 and the above curing reactivity comparison was

proved to be reliable. It was noted that all the samples showed

an activation energy a little higher than the typical value for

benzoxazine polycondensation (130−90 kJ/mol). The reason

might be that, besides the oxazine ring-opening reaction, the

furfural group participated in the side reaction, which led to a

higher observed activation energy.

In Ishida’s work,³⁰the higher curing reactivity of ortho-

amide functional benzoxazine was attributed to the accel-

I

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Scheme 4. Illustration for How the Intramolecular Hydrogen Bonding and Its Strength Inﬂuence the Curing of Benzoxazine

Precursors (o-AB-fbz, o-AF-fbz, and o-AP-fbz Set as an Example)

Figure 6. DSC heating curves for the cured benzoxazine networks (a). FT-IR spectra of all the cured benzoxazine networks before and after curing

reaction (b).

rings, and then reduced the catalytic eﬀects. However, for the

p-AB-fbz, all of the protons in NH groups had the opportunity

to participate in the formation of intermolecular H-bonding.

The strength of intermolecular H-bonding involved with the

oxygen in oxazine rings followed the order of p-AB-fbz > p-AF-

fbz > p-AP-fbz (Figure 3), which was the reason for the lowest

curing temperature of p-AB-fbz compared with p-AF-fbz and p-

AP-fbz.

As for the determination of curing condition for benzoxazine

precursors, the curing process should be started at a

temperature at which the curing reaction is relatively mild.

And the temperature for postcuring must be high enough to

ensure the full curing reaction in a proper curing time.

Additionally, the possible rearrangement under high temper-

ature should be controlled during the curing process.

According to previous publication,⁵⁵ortho-amide functional

polybenzoxazines might suﬀer rearrangement at around 250 °C

to form polybenzoxazole after the polybenzoxazines cured

completely. In this work, o-AB-fbz, o-AF-fbz, and o-AP-fbz also

possessed o-amide groups and the post rearrangement would

be inevitable at higher temperature, which was already

indicated by the broad peak around 250 °C in the DSC

curve for o-AB-fbz (Figure 4). The formation of polybenzox-

azole structure is not conducive to H-bonding investigation.

Therefore, the curing procedures for all the samples were

determined as follows: 2 h at 180 °C, 2 h at 200 °C, and 4 h at

220 °C according to the literature.⁵⁵Figure 6a shows the DSC

heating curves for all the systems after curing reaction. Before

250 °C, no obvious peaks were observed, which indicated the

complete curing reaction of oxazine rings. The small peaks

around 280 °C for poly(o-AB-fbz), poly(o-AF-fbz), and poly(o-

AP-fbz) were attributed to the rearrangement of polybenzox-

azine into polybenzoxazole according to published work.⁵⁵

FT-IR spectra were employed for further investigation of the

curing procedures. In Figure 6b, the typical absorption band at

925 cm⁻¹, mainly standing for oxazine ring,⁶⁶disappeared after

the monomers was cured into polybenzoxazine. And the

characteristic peak shown at 1362 cm⁻¹(CH₂wagging into the

closed benzoxazine ring) was also disappeared after curing

reaction. As for the band at 1595 cm⁻¹, indicating the

formation of polybenzoxazole structure, was not observed,

which proved that the rearrangement of poly(o-AB-fbz),

poly(o-AF-fbz) and poly(o-AP-fbz) during the curing reaction

was inhibited. The typical absorption bonds for benzoxazine

J

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Figure 7. DMA curves of cured systems: (a) storage modulus as a function of temperature and (b) tan δ as a function of temperature for the

diﬀerent cured resins; TGA curves (c) and diﬀerential curves (d) for the cured systems under N₂.

Table 2. Mechanical and Thermal Properties for the Diﬀerent Cured Systems

samples

T_g(°C) storage modulus at T_g+ 30 K (MPa) cross-link density (mol dm⁻³

)

T_d5%(°C) T_max(°C) char yield at 800 °C (%)

poly(o-AB-fbz)

poly(o-AF-fbz)

poly(o-AP-fbz)

poly(p-AB-fbz)

poly(p-AF-fbz)

poly(p-AP-fbz)

275

285

292

257

268

273

233

229

221

204

223

210

16.1

15.6

15.2

14.6

15.7

14.6

338

351

354

347

370

364

542

537

532

450

507

482

57

62

59

63

61

ring in p-AB-fbz, p-AF-fbz, and p-AP-fbz were presented at 928

and 1363 cm⁻¹. And their disappearance after curing reaction

suggested the formation of poly(p-AB-fbz), poly(p-AF-fbz),

and poly(p-AP-fbz) again. Combined with the results from

DSC and FT-IR, the report concluded that all of the model

benzoxazine precursors were cured completely, following the

programmed temperature rising method, and more important,

the rearrangement reaction was inhibited successfully.

The Thermal Performance of Polybenzoxazines. DMA

was taken to investigate the dynamic mechanical properties of

cured systems. Figure 7a shows the storage modulus, E′, of

poly(o-AB-fbz), poly(o-AF-fbz), poly(o-AP-fbz), poly(p-AB-

fbz), poly(p-AF-fbz), and poly(p-AP-fbz) as a function of

temperature, respectively. The E′ of cured polybenzoxazines

was in the range 2.6−2.9 GPa at room temperature and no

signiﬁcant diﬀerences were noticed. Conventionally, T_gof

cured samples was determined by the peak temperature in the

tan δ vs temperature graph (Figure 7(b)). And poly(o-AB-fbz),

poly(o-AF-fbz), poly(o-AP-fbz), poly(p-AB-fbz), poly(p-AF-

fbz), and poly(p-AP-fbz) showed T_gin the sequence 275, 285,

292, 257, 268, and 273 °C. Obviously, the T_gof cured

polybenzoxazines containing diﬀerent bridging units (benzene,

furan, and pyridine) followed the order poly(o-AB-fbz) <

poly(o-AF-fbz) < poly(o-AP-fbz) and poly(p-AB-fbz) < poly(p-

AF-fbz) < poly(p-AP-fbz). In addition, compared with the

literature results,^59,67−70these polybenzoxazines showed higher

T_grelative to the polybenzoxazines derived from bisphenol A,

whose T_gwas usually about 180 °C.

As we know, the thermal properties of polybenzoxazines will

be greatly aﬀected by their cross-link density, which is deﬁned

as the number of moles of network chains per volume unit of

the cured material.^7,71And the rubber elasticity theory is

usually applied for the determination of thermosets’ cross-link

density. According to this theory, the storage modulus in the

rubbery plateau (E′_R) will change with the cross-link density

and it can be calculated by the following eq 3:

E′_R

3RT

ϑ =

c

(3)

where ϑ_cwas the cross-link density (mol dm⁻³), E′_Rwas the

plateau modulus in the rubbery state (Pa), which was read at

T_g+ 30 K, R was the gas constant (J mol⁻¹g⁻¹), and T was the

absolute temperature at T_g+ 30 K. The thus calculated ϑ_c

values for all the cured systems were listed in Table 2. It was

noted that all the polybenzoxazines showed similar cross-link

K

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Figure 8. Normalized FT-IR spectra of all the cured diﬀerent systems at various temperatures (25, 50, 100, 150, 200, and 250 °C): (a) poly(o-AB-

fbz), (b) poly(o-AF-fbz), (c) poly(o-AP-fbz), (d) poly(p-AB-fbz), (e) poly(p-AF-fbz), and (f) poly(p-AP-fbz).

density, which indicated that the diﬀerence in T_gwas not

caused by the cross-link density.

higher than those of o-amide substituted analogues. As for the

fastest degradation temperature (T_max), T_maxof poly(p-AB-

fbz), poly(p-AF-fbz), and poly(p-AP-fbz) was in turn 457, 473,

and 488 °C. However, the T_maxof poly(o-AB-fbz), poly(o-AF-

fbz) and poly(o-AP-fbz) was high up to 542, 537, and 532 °C,

respectively, which was much higher than those of the

analogues with para-substituted amide group. This phenom-

enon was related to the fact that there were two peaks in the

diﬀerential TGA curves of poly(o-AB-fbz), poly(o-AF-fbz), and

poly(o-AP-fbz) (Figure 7d), while only one peak was observed

in the diﬀerential curves of poly(p-AB-fbz), poly(p-AF-fbz),

and poly(p-AP-fbz). These results conﬁrmed the conversion of

o-amide functionalized polybenzoxazines to polybenzoxazole

again, which was accompanied by the release of water. In the

Supporting Information, the conversion reaction was clearly

identiﬁed with the help of DSC, FT-IR and TGA (Figure S21).

As far as the char yield was concerned, it was as high as 57 wt

% for poly(o-AB-fbz), 62 wt % for poly(o-AF-fbz), 62 wt % for

poly(o-AP-fbz), 59 wt % for poly(p-AB-fbz), 63 wt % for

poly(p-AF-fbz), and 61 wt % for poly(p-AP-fbz) at 800 °C.

These results were normal.

Considering the similar structures of synthesized model

monomers and literature results,⁷²⁻⁷⁴the diﬀerent T_gof the

cured polybenzoxazines was possibly caused by the formation

and varied strength of H-bonding, which would be clearly

analyzed and explained in the following part. What should be

mentioned here was that the tan δ of poly(o-AB-fbz), poly(o-

AF-fbz) or poly(o-AP-fbz) did not show a dramatic decrease

after the peak temperature with a indistinct broad peak,

presented in Figure 7b. It might be caused by the

cyclodehydration of o-amide-functionalized polybenzoxazine

above T_g, after which the polybenzoxazine was converted into

polybenzoxazole. These results were consistent with the DSC

heating curves in Figure 6a, which were also reported in

Ishida’s work.⁵⁵

The thermal stability of the cured system was investigated by

TGA under nitrogen (Figure 7(c)) and the related data was

listed in Table 2. The temperature of 5% weight loss (T_d5%

)

was 338 °C for poly(o-AB-fbz), 351 °C for poly(o-AF-fbz), 354

°C for poly(o-AP-fbz), 347 °C for poly(p-AB-fbz), 370 °C for

poly(p-AF-fbz), and 364 °C for poly(p-AP-fbz). In general, the

T_d5%values of polybenzoxazines with p-amide structure were

FT-IR Spectra Analysis for the Cured Systems. As

mentioned above, the high T_gof polybenzoxazines in this work

L

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Figure 9. Diﬀerent types of hydrogen bonding in all the cured systems depicted in FT-IR spectra at 25 (a) and 250 °C (b) after normalization;

splitting FT-IR spectra between 2650 and 3700 cm⁻¹of poly(o-AF-fbz) and poly(p-AF-fbz) under 25 (c) or 250 °C (d); (e) comparison of split

peaks’ integrated areas in diﬀerent systems at 25 and 250 °C.

was possibly related to the formation of H-bonding and their

diﬀerence was caused by the varied H-bonding strength.

Normalized FT-IR spectra were used to investigate the H-

bonding interactions in all the cured systems at various

temperatures (25, 50, 100, 150, 200, and 250 °C) (Figure 8).

As we know, the intramolecular H-bonding -O⁻--H⁺N−,

−OH--N−, −OH--O− and −OH--π usually show the

characteristic absorption bonds at ∼2900, ∼ 3180, ∼ 3460

and ∼3550 cm⁻¹, respectively. The intermolecular H-bonding

−OH--N− and −OH--O− can be identiﬁed by the typical

absorption bands at ∼3300 and ∼3380 cm⁻¹. And the peaks

above 3600 cm⁻¹refer to the free − OH and intermolecular

−OH--π.^39,40,75,76In order to clearly study the features of H-

bonding in the cured systems, only the absorption peaks

between 2650 and 3700 cm⁻¹were shown. In Figure 8a−f, it

was noted that the typical absorption bonds for H-bonding was

gradually deceased when the temperature was increased from

25 to 250 °C and the still presence of H-bonding interactions

in poly(o-AB-fbz, poly(o-AF-fbz), poly(o-AP-fbz), poly(p-AB-

fbz), poly(p-AF-fbz), and poly(p-AP-fbz) at 250 °C was

supported by the obvious characteristic absorption bonds. This

result meant that it was reasonable for us to attribute the

relatively higher T_gto the formation of H-bonding even at 250

°C, which hindered the motion of molecular segments and

then led to a higher T_g. However, what type of H-bonding,

intermolecular or intramolecular (“In the same or diﬀerent

chain/unit” might be more accurate in the cross-linked

systems. However, considering the expression custom,

“intermolecular or intramolecular” was still employed here

for the cured resins), tend to exist at higher temperature and

how would the chemical structure (diﬀerent bridging units:

benzene, furan and pyridine) inﬂuence the formation of inter-

or intramolecular H-bonding in the cured resins, were

remained to be answered in the following part.

Figure 9 showed the FT-IR spectra comparison between all

the cured systems at 25 (Figure 9a) and 250 °C (Figure 9b).

On the basis of the chemical structures of cured systems, the

characteristic bonds showing at 2925 cm⁻¹were assigned to

the intramolecular H-bonding RCON−H--N (pyridine) and

RCON−H--O (furan or Ar). While the peaks centered at 3120

cm⁻¹were corresponding to the intramolecular H-bonding of

ArO−H--N (pyridine or N atom in other groups), peaks

around 3350 cm⁻¹were standing for the inter- or intra-

molecular H-bonding ArOH--O (carbonyl, furan or Ar) and

intermolecular H-bonding ArOH--N, peaks showing at 3600

cm⁻¹were attributed to the intramolecular H-bonding −OH--

M

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Figure 10. Molecular structures of cross-linked polybenzoxazines: (a) for poly(o-AB-fbz), poly(o-AF-fbz), and poly(o-AP-fbz) and (b) for poly(p-

AB-fbz), poly(p-AF-fbz), and poly(p-AP-fbz).

Table 3. Simulation Properties for Cross-Linked Polybenzoxazines

samples

poly(o-AB-fbz)

poly(o-AF-fbz)

poly(o-AP-fbz)

poly(p-AB-fbz)

poly(p-AF-fbz)

poly(p-AP-fbz)

density @ 300 K (g/cm³)

1.13

1.15

1.13

T_g(°C)

267

3.59 0.11

277

3.98 0.12

289

4.04 0.08

240

3.19 0.08

252

3.43 0.11

265

3.49 0.09

H-bonding density (10⁻³mol/cm³)

π (benzene, furan or pyridine) and free −OH. It was noted

that, in general, the intensity of all the absorption bonds

between 2650 and 3700 cm⁻¹was in accordance with the

following order: poly(o-AP-fbz) > poly(o-AF-fbz) > poly(o-

AB-fbz) > poly(p-AP-fbz) > poly(p-AF-fbz) > poly(p-AB-fbz),

both at 25 and 250 °C, which roughly agreed with the order of

their T_g. In order to obtain deeper information, the FT-IR

spectra were analyzed by means of overlapping peak resolving

(keeping the characteristic frequency at 2925, 3120, 3350, and

3600 cm⁻¹, where all of the R²values of the ﬁtting curves are

above 0.9). For conciseness, only the split spectra of poly(o-

AF-fbz) and poly(p-AF-fbz) under 25 (Figure 9c) and 250 °C

(Figure 9d) were shown here and the others were put in the

Supporting Information (Figures S22−S29). According to the

peak assignment, the FT-IR spectra at 25 °C was split into four

peaks, respectively centered at 2925, 3120, 3350, and 3600

cm⁻¹, and the spectra at 250 °C were divided into three parts

(2925, 3120, and 3350 cm⁻¹) due to the disappearance of

absorption at 3600 cm⁻¹. It was easy to notice that the

intensity of split peaks were varied in the diﬀerent systems and

their integrated areas were taken to qualitatively compare the

strength of diﬀerent H-bonding interactions. The integrated

areas of all the split peaks in diﬀerent systems were obtained

and their ratios were demonstrated in Figure 9e. At 25 °C, the

inter- and intramolecular H-bonding ArOH--O (carbonyl,

furan or Ar) and intermolecular H-bonding ArOH--N were

predominant in all the cured systems. Compared with the p-

amide-substituted systems, not only the strength of H-bonding

in poly(o-AP-fbz), poly(o-AF-fbz) and poly(o-AB-fbz), but also

the percentage of intramolecular H-bonding, such as RCON−

H--N (pyridine), RCON−H--O (furan or Ar), ArOH--N

(pyridine or other groups), and ArOH--O (carbonyl), were

relatively higher. When the temperature was increased up to

250 °C, the intensity of absorption bonds in all the systems

were dramatically decreased, especially the ones standing for

the intermolecular H-bonding, although the IR extinction

coeﬃcient of even the same band will change at diﬀerent

temperature, which will inﬂuence the intensity of the

absorption bonds. The temperature-dependent IR is usually

employed to semiquantitatively compare the strength of H-

bonding, considering the slight change of extinction coeﬃcient

at diﬀerent temperature. In this work, the decreased intensity

of absorption bonds at higher temperature should be related

with the decreased H-bonding strength. And this was easy to

understand because the strength of intramolecular H-bond was

usually stronger than the interaction between diﬀerent

molecules, and the higher temperature would easily destroy

weaker H-bonding, then led to the increased ratio of

intramolecular ones. As mentioned above, the bridging units

in diﬀerent benzoaxzines, benzene, furan, and pyridine

demonstrated diﬀerent electron-donating abilities, which

could inﬂuence the formation and strength of H-bonding in

the cured resins. The pyridine units showed the strongest

ability to attract protons and was followed by the furan units.

Therefore, the density or strength of H-bonding in the o-

amide-functionalized systems should follow the order poly(o-

AP-fbz) > poly(o-AF-fbz) > poly(o-AB-fbz). This speculation

was supported by the H-bonding strength comparison in

Figure 9e and in line with the order of their T_g. As for the p-

amide substituted systems, their T_gfollowed the order of

poly(p-AP-fbz) > poly(p-AF-fbz) > poly(p-AB-fbz). And the

reason was also related to the H-bonding strength. In addition,

compared to the strength of intramolecular H-bonding with

the intermolecular interaction at 25 and 250 °C, the

intramolecular H-bonding was undoubtedly playing a more

important role in aﬀecting the T_gof cured resins, due to the

lower stability of intramolecular hydrogen bonds at high

temperature. On the basis of these results, the conclusion

might be drawn that there was a close positive correlation

between the H-bonding interactions, especially the intra-

molecular ones, and thermal property (T_g) of polybenzox-

azines.

MD Simulation for the Model Systems. After the

multistep cross-linking reaction and topology relaxation via

MD simulation, the ﬁnal structures of polybenzoxazines were

shown in Figure 10. The density at room temperature (300 K)

was 1.13, 1.15, 1.15, 1.13, 1.13, and 1.13 g cm⁻³for poly(o-AB-

fbz), poly(o-AF-fbz), poly(o-AP-fbz), poly(p-AB-fbz), poly(p-

AF-fbz), and poly(p-AP-fbz), respectively. And the relevant

data is summarized in Table 3.

The well-relaxed network of each polybenzoxazine system

was obtained after being experienced an equilibration at 850 K

N

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

Figure 11. Temperature dependence of density for polybenzoxazines (a, b) and the H-bonding length distribution (c) for poly(o-AB-fbz), poly(o-

AF-fbz), and poly(o-AP-fbz) and (d) for poly(p-AB-fbz), poly(p-AF-fbz), and poly(p-AP-fbz).

in NPT, followed by an annealing simulation, i.e. a cooling

down process at the rate of 50 K/200 ps to 250 K using NPT-

MD simulations. According to the results from the annealing

simulation, the dependence of density on the temperature for

each cross-linked system was shown in Figure 11, parts a and b.

The characteristic turning points in the slope of the density−

temperature curve was observed, where the temperature was

corresponding to T_g. In order to get the exact values of T_g,

linear ﬁt with more than 98% conﬁdence was performed and

the thus obtained T_gs were listed in Table 3. As shown in

Table 3, the calculated T_gs for the cross-linked systems were

267 °C for poly(o-AB-fbz), 277 °C for poly(o-AF-fbz), 289 °C

for poly(o-AP-fbz), and 240 °C for poly(p-AB-fbz), 252 °C for

poly(p-AF-fbz), 265 °C for poly(p-AP-fbz), respectively, which

were all a light lower than the experimentally observed T_gs

(275 °C for poly(o-AB-fbz), 285 °C for poly(o-AF-fbz), 292

°C for poly(o-AP-fbz), and 257 °C for poly(p-AB-fbz), 268 °C

for poly(p-AF-fbz), 273 °C for poly(p-AP-fbz) based on above

DMA measurement). The reason might be that the molecular

dynamic simulation was conducted only based on the oxazine

ring polymerization mechanism disregarding the possible

furfuryl group polymerization. Actually, the furfuryl group

would participate in the polymerization in various mechanisms,

which led to the elevated T_gs.

bonding strength is related to the bond length. Therefore, the

distribution of H-bonds length for all the cross-linked

polybenzoxazine was summarized and represented in Figure

11, parts c and d. In comparison of the systems containing o-

amide groups (poly(o-AB-fbz), poly(o-AF-fbz), and poly(o-AP-

fbz)) with the para-amide functionalized systems (poly(p-AB-

fbz), poly(p-AF-fbz), and poly(p-AP-fbz)), more hydrogen

bonds with the length less than 0.2 nm were observed in the

former systems, indicating their higher hydrogen bond

strength. As for poly(o-AB-fbz), poly(o-AF-fbz), and poly(o-

AP-fbz), the content of hydrogen bonds showing the bond

length in the range of 0.15 to 0.2 nm followed the order of

poly(o-AB-fbz) < poly(o-AF-fbz) < poly(o-AP-fbz). This was

same as the sequence of their T_gs, either obtained from DSC

measurement or MD simulation. In Figure 11d, it could note

that the statistical length of hydrogen bonds in poly(p-AP-fbz)

was shorter than those in poly(p-AB-fbz) and poly(p-AF-fbz),

although the length distribution of hydrogen bonds was more

even compared with that in Figure 11c. On the basis of these

results, it could conclude that both the overall H-bonding

strength and T_gobtained from MD simulation followed the

order of poly(o-AB-fbz) < poly(o-AF-fbz) < poly(o-AP-fbz)

and poly(p-AB-fbz) < poly(p-AF-fbz) < poly(p-AP-fbz), which

was same as the experimental results and veriﬁed the

conclusion based on experiments.

As deﬁned by Kim and Mattice,⁵²a H-bonding formed when

the distance between a donor −OH and an acceptor -O or -N

was less than 0.25 nm and the angle of − O−H−O− or − O−

H−N was in the range of 120° to 180°. In this work, the values

of H-bonding density were averaged over ﬁve frames in the last

100 ps during the MD simulations. From Table 3, it was noted

that the calculated H-bonding density was related to the values

of T_g. The higher the H-bonding density, the higher the T_g. As

a matter of fact, not only the quantity, but also the strength of

each bond inﬂuence the overall strength of H-bonding. And H-

CONCLUSION

■

The model benzoxazine monomers o-AB-fbz, o-AF-fbz, o-AP-

fbz, p-AB-fbz, p-AF-fbz, and p-AP-fbz were carefully designed

and successfully synthesized. It was found that not only the

formation but also the type and strength of H-bonding played

a signiﬁcant role in determining the curing behaviors of

benzoxazines. It was the H-bonding involved with the oxygen

atom in oxazine ring, rather than all of the H-bonding, that

O

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

(4) Takeichi, T.; Kawauchi, T.; Agag, T. High Performance

Polybenzoxazines as a Novel Type of Phenolic Resin. Polym. J.

2008, 40, 1121−1131.

accelerated the curing reactions, and the higher strength will

lead to higher reactivity. Diﬀerent from the curing reaction, it

was the overall H-bonding interactions that were important,

including all the inter- and intramolecular ones, and their

strengths were positively correlated with the T_gof poly-

benzoxazines based on the FT-IR analysis. However, the

intramolecular H-bonding played a more important role in

aﬀecting the T_g, due to their higher stabilities at high

temperature. It was the ﬁrst time that not only the formation

of H-bonding but also their type and strength in benzoxazine

monomers and the cured products were characterized and

qualitatively investigated. Their roles in aﬀecting the curing

behaviors of benzoxazine and thermal properties of cured

polybenzoxazines were studied in detail. The information

provided in this paper could help us understand polybenzox-

azines more deeply and manipulate their properties eﬃciently.

(5) Holly, F. W.; Cope, A. C. Condensation Product of Aldehydes

and Ketones with o-Aminobenzyl Alcohol and o-Hydroxybenzyl-

amine. J. Am. Chem. Soc. 1944, 66, 1875−1879.

(6) Ning, X.; Ishida, H. Phenolic Materials via Ring-Opening

Polymerization-Synthesis and Characterization of Bisphenol-A Based

Benzoxazines and Their Polymers. J. Polym. Sci., Part A: Polym. Chem.

1994, 32, 1121−1129.

(7) Ishida, H.; Allen, D. J. Mechanical Characterization of

Copolymers Based on Benzoxazine and Epoxy. Polymer 1996, 37,

4487−4495.

(8) Takeichi, T.; Saito, Y.; Agag, T.; Muto, H.; Kawauchi, T. High-

Performance Polymer Alloys of Polybenzoxazine and Bismaleimide.

Polymer 2008, 49, 1173−1179.

(9) Takeichi, T.; Guo, Y.; Agag, T. Synthesis and Characterization of

Poly(urethane-benzoxazine) Films as Novel Type of Polyurethane/

Phenolic Resin Composites. J. Polym. Sci., Part A: Polym. Chem. 2000,

38, 4165−4176.

ASSOCIATED CONTENT

* Supporting Information

■

S

(10) Wang, M. W.; Jeng, R. J.; Lin, C. H. Origin of The Rapid

Trimerization of Cyanate Ester in a Benzoxazine/Cyanate Ester

Blend. Macromolecules 2015, 48, 2417−2421.

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.macro-

mol.8b00741.

(11) Li, X. D.; Gu, Y. The Co-Curing Process of a Benzoxazine-

Cyanate System and The Thermal Properties of The Copolymers.

Polym. Chem. 2011, 2, 2778−2781.

NMR spectra of compounds 1−6, o-AB-fbz, o-AF-fbz, o-

AP-fbz, p-AB-fbz, p-AF-fbz, and p-AP-fbz; mass spectra

of o-AB-fbz, o-AF-fbz, o-AP-fbz, p-AB-fbz, p-AF-fbz, and

p-AP-fbz; and splitting FT-IR spectra for poly(o-AB-

fbz), poly(o-AF-fbz), poly(o-AP-fbz), poly(p-AB-fbz),

poly(p-AF-fbz), and poly(p-AP-fbz) at 25 and 250 °C,

respectively (PDF)

(12) Ishida, H.; Chaisuwan, T. Mechanical Property Improvement of

Carbon Fiber Reinforced Polybenzoxazine by Rubber Interlayer.

Polym. Compos. 2003, 24, 597−607.

(13) Zeng, M.; Wang, J.; Li, R. R.; Liu, J. X.; Chen, W.; Xu, Q. Y.;

Gu, Y. The Curing Behavior and Thermal Property of Graphene

Oxide/Benzoxazine Nanocomposites. Polymer 2013, 54, 3107−3116.

(14) Santhosh Kumar, K. S.; Reghunadhan Nair, C. P.; Ninan, K. N.

Silica Fiber-Polybenzoxazine-Syntactic Foams; Processing and Prop-

erties. J. Appl. Polym. Sci. 2008, 107, 1091−1099.

AUTHOR INFORMATION

Corresponding Author

*Address: 1219 Zhongguan West Rd., Zhenhai District,

Ningbo 315201, China. E-mail: liuxq@nimte.ac.cn. (X.L.).

■

(15) Ishida, H.; Low, H. Y. Synthesis of Benzoxazine Functional

Silane and Adhesion Properties of Glass-Fiber-Reinforced Poly-

benzoxazine Composites. J. Appl. Polym. Sci. 1998, 69, 2559−2567.

(16) Yang, C. C.; Lin, Y. C.; Wang, P. I.; Liaw, D. J.; Kuo, S. W.

Polybenzoxazine/Single-Walled Carbon Nanotube Nanocomposites

Stabilized Through Noncovalent Bonding Interactions. Polymer 2014,

55, 2044−2050.

(17) Agag, T.; Takeichi, T. Synthesis, Characterization and Clay-

Reinforcement of Epoxy Cured with Benzoxazine. High Perform.

Polym. 2002, 14, 115−132.

ORCID

Xiaoqing Liu: 0000-0002-7417-1326

Shiyu Du: 0000-0001-6707-3915

Author Contributions

^#This author made an equal contribution to the ﬁrst author.

(18) Agag, T.; Takeichi, T. Preparation and Cure Behavior of

Organoclay-Modified Allyl-Functional Benzoxazine Resin and The

Properties of Their Nanocomposites. Polym. Compos. 2008, 29, 750−

757.

(19) Andreu, R.; Reina, J. A.; Ronda, J. C. Carboxylic Acid-

Containing Benzoxazines as Efficient Catalysts in The Thermal

Polymerization of Benzoxazines. J. Polym. Sci., Part A: Polym. Chem.

2008, 46, 6091−6101.

(20) Agag, T.; Takeichi, T. Novel Benzoxazine Monomers

Containing p-Phenyl Propargyl Ether: Polymerization of Monomers

and Properties of Polybenzoxazines. Macromolecules 2001, 34, 7257−

7263.

(21) Zhang, K.; Liu, J.; Ishida, H. An Ultrahigh Performance Cross-

Linked Polybenzoxazole via Thermal Conversion from Poly-

(benzoxazine amic acid) Based on Smart o-Benzoxazine Chemistry.

Macromolecules 2014, 47, 8674−8681.

(22) Zhang, K.; Ishida, H. An Anomalous Trade-Off Effect on The

Properties of Smart ortho-Functional Benzoxazines. Polym. Chem.

2015, 6, 2541−2550.

(23) Zhang, K.; Han, L.; Froimowicz, P.; Ishida, H. A Smart Latent

Catalyst Containing o-Trifluoroacetamide Functional Benzoxazine:

Precursor for Low Temperature Formation of Very High Performance

Polybenzoxazole with Low Dielectric Constant and High Thermal

Stability. Macromolecules 2017, 50, 6552−6560.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

The authors greatly thank the National Key Technology

Support Program (2015BAD15B08), the National Key

Research and Development Program of China

(2018YFD0400700), the National Natural Science Foundation

of China (NSFC 51373194; NSFC 51503217), the Chinese

Academy of Sciences (Key Research Program ZDRW-CN-

2016-1), the Research Project of Technology Application for

Public Welfare of Zhejiang Province (2017C31081), and the

Ningbo Natural Science Foundation (No. 2017A610053) for

ﬁnancial support.

REFERENCES

■

(1) Ishida, H. Handbook of Benzoxazine Resins 2011, 3−81.

(2) Ghosh, N. N.; Kiskan, B.; Yagci, Y. Polybenzoxazines-New High

Performance Thermosetting Resins: Synthesis and Properties. Prog.

Polym. Sci. 2007, 32, 1344−1391.

(3) Nair, C. P. R. Advances in Addition-Cure Phenolic Resins. Prog.

Polym. Sci. 2004, 29, 401−498.

P

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

(24) Hanbeyoglu, B.; Kiskan, B.; Yagci, Y. Hydroxyl Functional

Polybenzoxazine Precursor as A Versatile Platform for Post-Polymer

Modifications. Macromolecules 2013, 46, 8434−8440.

(25) Agag, T.; Geiger, S.; Alhassan, S. M.; Qutubuddin, S.; Ishida, H.

Low-Viscosity Polyether-Based Main-Chain Benzoxazine Polymers:

Precursors for Flexible Thermosetting Polymers. Macromolecules

2010, 43, 7122−7127.

(26) Kudoh, R.; Sudo, A.; Endo, T. Synthesis of Eight-Membered

Lactone Having Tertiary Amine Moiety by Ring-Expansion Reaction

of 1,3-Benzoxazine and Its Anionic Ring-Opening Polymerization

Behavior. Macromolecules 2009, 42, 2327−2329.

(27) Froimowicz, P.; Zhang, K.; Ishida, H. Intramolecular Hydrogen

Bonding in Benzoxazines: When Structural Design Becomes Func-

tional. Chem. - Eur. J. 2016, 22, 2691−2707.

(28) Baqar, M.; Agag, T.; Huang, R. Z.; Maia, J.; Qutubuddin, S.;

Ishida, H. Mechanistic Pathways for The Polymerization of Methylol-

Functional Benzoxazine Monomers. Macromolecules 2012, 45, 8119−

8125.

(29) Zhang, K.; Froimowicz, P.; Han, L.; Ishida, H. Hydrogen-

Bonding Characteristics and Unique Ring-Opening Polymerization

Behavior of Ortho-Methylol Functional Benzoxazine. J. Polym. Sci.,

Part A: Polym. Chem. 2016, 54, 3635−3642.

(30) Han, L.; Zhang, K.; Ishida, H.; Froimowicz, P. Study of The

Effects of Intramolecular and Intermolecular Hydrogen-Bonding

Systems on The Polymerization of Amide-Containing Benzoxazines.

Macromol. Chem. Phys. 2017, 218, 1600562.

(31) Kolanadiyil, S. N.; Minami, M.; Endo, T. Synthesis and

Thermal Properties of Difunctional Benzoxazines with Attached

Oxazine Ring at the Para-, Meta-, and Ortho-Position. Macromolecules

2017, 50, 3476−3488.

(32) Arslan, M.; Kiskan, B.; Yagci, Y. Benzoxazine-Based Thermosets

with Autonomous Self-Healing Ability. Macromolecules 2015, 48,

1329−1334.

(33) Taskin, O. S.; Kiskan, B.; Yagci, Y. Polybenzoxazine Precursors

As Self-Healing Agents for Polysulfones. Macromolecules 2013, 46,

8773−8778.

(34) Alhwaige, A. A.; Agag, T.; Ishida, H.; Qutubuddin, S. Biobased

Chitosan/Polybenzoxazine Cross-Linked Films: Preparation in

Aqueous Media and Synergistic Improvements in Thermal and

Mechanical Properties. Biomacromolecules 2013, 14, 1806−1815.

(35) Shen, X.; Dai, J.; Liu, Y.; Liu, X.; Zhu, J. Synthesis of High

Performance Polybenzoxazine Networks from Bio-Based Furfuryl-

amine: Furan vs Benzene Ring. Polymer 2017, 122, 258−269.

(36) Lin, C. H.; Chang, S. L.; Shen, T. Y.; Shih, Y. S.; Lin, H. T.;

Wang, C. F. Flexible Polybenzoxazine Thermosets with High Glass

Transition Temperatures and Low Surface Free Energies. Polym.

Chem. 2012, 3, 935−945.

(37) Sini, N. K.; Endo, T. Toward Elucidating the Role of Number

of Oxazine Rings and Intermediates in The Benzoxazine Backbone on

Their Thermal Characteristics. Macromolecules 2016, 49, 8466−8478.

(38) Froimowicz, P.; Rodriguez Arza, C.; Ohashi, S.; Ishida, H.

Tailor-Made and Chemically Designed Synthesis of Coumarin-

Containing Benzoxazines and Their Reactivity Study Toward Their

Thermosets. J. Polym. Sci., Part A: Polym. Chem. 2016, 54, 1428−

1435.

(43) Che, P. H.; Lu, F.; Nie, X.; Huang, Y. Z.; Yang, Y. L.; Wang, F.;

Xu, J. Hydrogen Bond Distinction and Activation Upon Catalytic

Etherification of Hydroxyl Compounds. Chem. Commun. 2015, 51,

1077−1080.

(44) Wang, L. X.; Hu, B. Q.; Xiang, J. F.; Cui, J.; Hao, X.; Liang, T.

L.; Tang, Y. L. N-Aryl-Substituted Anthranilamides with Intra-

molecular Hydrogen Bonds. Tetrahedron 2014, 70, 8588−8591.

(45) Pearson, A. J.; Panda, S. N-Prolinylanthranilamide Pseudopep-

tides as Bifunctional Organocatalysts for Asymmetric Aldol Reactions.

Org. Lett. 2011, 13, 5548−5551.

(46) Yarovsky, I.; Evans, E. Computer Simulation of Structure and

Properties of Crosslinked Polymers: Application to Epoxy Resins.

Polymer 2002, 43, 963−969.

(47) Varshney, V.; Patnaik, S. S.; Roy, A. K.; Farmer, B. L. A

Molecular Dynamics Study of Epoxy-Based Networks: Cross-linking

Procedure and Prediction of Molecular and Material Properties.

Macromolecules 2008, 41, 6837−6842.

(48) Li, C.; Strachan, A. Molecular Simulations of Crosslinking

Process of Thermosetting Polymers. Polymer 2010, 51, 6058−6070.

(49) Li, C.; Strachan, A. Molecular Dynamics Predictions of

Thermal and Mechanical Properties of Thermoset Polymer

EPON862/DETDA. Polymer 2011, 52, 2920−2928.

(50) Masoumi, S.; Arab, B.; Valipour, H. A Study of Thermo-

Mechanical Properties of The Cross-linked Epoxy: An Atomistic

Simulation. Polymer 2015, 70, 351−360.

(51) Saiev, S.; Bonnaud, L.; Dubois, P.; Beljonne, D.; Lazzaroni, R.

Modeling The Formation and Thermomechanical Properties of

Polybenzoxazine Thermosets. Polym. Chem. 2017, 8, 5988−5999.

(52) Kim, W. K.; Mattice, W. L. A Fully Atomistic Model of An

Amorphous Polybenzoxazine at Bulk Density. Comput. Theor. Polym.

Sci. 1998, 8, 353−361.

(53) Kim, W. K.; Mattice, W. L. Static and Dynamic Behavior of

H₂O and O₂Penetrants in A Polybenzoxazine. Macromolecules 1998,

31, 9337−9344.

(54) Gomes, M.; Gandini, A.; Silvestre, A. J. D.; Reis, B. Synthesis

and Characterization of Poly(2,5-furan dicarboxylate)s Based on A

Variety of Diols. J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 3759−

3768.

(55) Agag, T.; Liu, J.; Graf, R.; Spiess, H. W.; Ishida, H. Benzoxazole

Resin: A Novel Class of Thermoset Polymer via Smart Benzoxazine

Resin. Macromolecules 2012, 45, 8991−8997.

(56) Mayo, S. L.; Olafson, B. D.; Goddard, W. A. Dreiding - A

Generic Force-Field For Molecular Simulations. J. Phys. Chem. 1990,

94, 8897−8909.

(57) Gasteiger, J.; Marsili, M. Iterative Partial Equalization of Orbital

Electronegativity - A Rapid Access to Atomic Charges. Tetrahedron

1980, 36, 3219−3228.

(58) Samoletov, A. A.; Dettmann, C. P.; Chaplain, M. A. J.

Thermostats for ″Slow″ Configurational Modes. J. Stat. Phys. 2007,

128, 1321−1336.

(59) Froimowicz, P.; Paez, J. I.; Gerbino, L.; Ali, S. N.; Belgacem, M.

N.; Gandini, A.; Strumia, M. C. Effect of The Molecular Structure on

The Reactivity in A Family of Tetra-Amine Compounds Derived from

Jeffamines. Macromol. Res. 2012, 20, 800−809.

(60) Zhang, W. F.; Froimowicz, P.; Arza, C. R.; Ohashi, S.; Xin, Z.;

Ishida, H. Latent Catalyst-Containing Naphthoxazine: Synthesis and

Effects on Ring-Opening Polymerization. Macromolecules 2016, 49,

7129−7140.

(61) Ishida, H.; Rodriguez, Y. Curing Kinetics of A New

Benzoxazine-Based Phenolic Resin by Differential Scanning Calorim-

etry. Polymer 1995, 36, 3151−3158.

(62) Agag, T.; Arza, C. R.; Maurer, F. H. J.; Ishida, H. Primary

Amine-Functional Benzoxazine Monomers and Their Use for Amide-

Containing Monomeric Benzoxazines. Macromolecules 2010, 43,

2748−2758.

(39) Kim, H. D.; Ishida, H. A Study on Hydrogen Bonding in

Controlled-Structure Benzoxazine Model Oligomers. Macromol. Symp.

2003, 195, 123−146.

(40) Kim, H. D.; Ishida, H. A Study on Hydrogen-Bonded Network

Structure of Polybenzoxazines. J. Phys. Chem. A 2002, 106, 3271−

3280.

(41) Kim, H. D.; Ishida, H. Model Compounds Study on The

Network Structure of Polybenzoxazines. Macromolecules 2003, 36,

8320−8329.

́

(42) Dolensky, B.; Konvalinka, R.; Jakubek, M.; Kral, V.

(63) Kissinger, H. E. Reaction Kinetics in Differential Thermal

Analysis. Anal. Chem. 1957, 29, 1702−1706.

(64) Ozawa, T. A New Method of Analyzing Thermogravimetric

Data. Bull. Chem. Soc. Jpn. 1965, 38, 1881−1886.

Identification of Intramolecular Hydrogen Bonds as The Origin of

Malfunctioning of Multitopic Receptors. J. Mol. Struct. 2013, 1035,

124−128.

Q

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

Article

(65) Liu, Y.; Wang, J.; Xu, S. Synthesis and Curing Kinetics of

Cardanol-Based Curing Agents for Epoxy Resin by In Situ

Depolymerization of Paraformaldehyde. J. Polym. Sci., Part A: Polym.

Chem. 2014, 52, 472−480.

(66) Han, L.; Iguchi, D.; Gil, P.; Heyl, T. R.; Sedwick, V. M.; Arza,

C. R.; Ohashi, S.; Lacks, D. J.; Ishida, H. Oxazine Ring-Related

Vibrational Modes of Benzoxazine Monomers Using Fully Aromati-

cally Substituted, Deuterated, ¹⁵N Isotope Exchanged, and Oxazine-

Ring-Substituted Compounds and Theoretical Calculations. J. Phys.

Chem. A 2017, 121, 6269−6282.

(67) Agag, T.; Jin, L.; Ishida, H. A New Synthetic Approach for

Difficult Benzoxazines: Preparation and Polymerization of 4,4

′-Diaminodiphenyl Sulfone-Based Benzoxazine Monomer. Polymer

2009, 50, 5940−5944.

(68) Zhang, K.; Liu, J.; Ohashi, S.; Liu, X. Y.; Han, Z. W.; Ishida, H.

Synthesis of High Thermal Stability Polybenzoxazoles via Ortho-

Imide-Functional Benzoxazine Monomers. J. Polym. Sci., Part A:

Polym. Chem. 2015, 53, 1330−1338.

(69) Su, Y. C.; Chen, W. C.; Chang, F. C. Preparation and

Characterization of Poly Seudorotaxanes Based on Adamantane-

Modified Polybenzoxazines and Beta-Cyclodextrin. Polymer 2005, 46,

1617−1623.

(70) Wang, J.; Wu, M. Q.; Liu, W. B.; Yang, S. W.; Bai, J. W.; Ding,

Q. Q.; Li, Y. Synthesis, Curing Behavior and Thermal Properties of

Fluorene Containing Benzoxazines. Eur. Polym. J. 2010, 46, 1024−

1031.

(71) Grishchuk, S.; Schmitt, S.; Vorster, O. C.; Karger-Kocsis, J.

Structure and Properties of Amine-Hardened Epoxy/Benzoxazine

Hybrids: Effect of Epoxy Resin Functionality. J. Appl. Polym. Sci.

2012, 124, 2824−2837.

(72) Liu, Y. L.; Chou, C. I. High Performance Benzoxazine

Monomers and Polymers Containing Furan Groups. J. Polym. Sci.,

Part A: Polym. Chem. 2005, 43, 5267−5282.

(73) Hu, F. S.; La Scala, J. J.; Sadler, J. M.; Palmese, G. R. Synthesis

and Characterization of Thermosetting Furan-Based Epoxy Systems.

Macromolecules 2014, 47, 3332−3342.

(74) Shen, X. B.; Liu, X. Q.; Dai, J. Y.; Liu, Y.; Zhang, Y. J.; Zhu, J.

How Does the Hydrogen Bonding Interaction Influence the

Properties of Furan-Based Epoxy Resins. Ind. Eng. Chem. Res. 2017,

56, 10929−10938.

(75) Skrovanek, D. J.; Howe, S. E.; Painter, P. C.; Coleman, M. M.

Hydrogen-Bonding in Polymers-Infrared Temperature Studies of An

Amorphous Polyamide. Macromolecules 1985, 18, 1676−1683.

(76) Wu, Q. J.; Liu, X. H.; Berglund, L. A. FT-IR Spectroscopic

Study of Hydrogen Bonding in PA6/Clay Nanocomposites. Polymer

2002, 43, 2445−2449.

R

DOI: 10.1021/acs.macromol.8b00741

Macromolecules XXXX, XXX, XXX−XXX

Article Doi

How Does the Hydrogen Bonding Interaction Influence the Properties of Polybenzoxazine? An Experimental Study Combined with Computer Simulation

DOI: 10.1021/acs.macromol.8b00741

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.macromol.8b00741

Full text of DOI:10.1021/acs.macromol.8b00741

Products guided by the article

R&D Labs maybe for 1245-91-6

Relevant to this article

Hot Product