Supporting Information for

Functionalized Graphene with Co-ZIF Adsorbed Borate Ions as an

Effective Flame Retardant and Smoke Suppression Agent for Epoxy

Resin

Wenzong Xu a, *, Xiaoling Wang a, Yun Wu b, Wu Li a, Chunying Chen a

a School of Materials Science and Chemical Engineering, Anhui Jianzhu University,

292 Ziyun Road, Hefei, Anhui 230601, China

b Institute of Science and Information Technology, Anhui University, Hefei, Anhui

230601, China

*Correspondence to: Wenzong Xu

(Tel./Fax:+86-0551-63828157. Email: wenzongxu@ahjzu.edu.cn)

S1

Calculation process of ICP experiment

In order to determine the amount of borate species in the resulting ZIF-67/RGO-

B, the Inductively Coupled Plasma (ICP) experiment was conducted, and the boron

content in ZIF-67/RGO-B was calculated and analyzed. The calculation basis is as

follows:

wtB%=m1/m*100% (1)

m1=m0-m2 (2)

---wtB% represents the mass percent of boron element in ZIF-67/RGO-B;

---m represents the mass of the product ZIF-67/RGO-B;

---m0 represents the mass of the boron element added before the reaction start;

---m1 represents the mass of the boron element adsorbed on the surface of

ZIF-67/RGO;

---m2 represents the mass of boron element that remains in the solution and is not

adsorbed at the end of the reaction.

Among them, m and m0 are known, m2 can be calculated according to the boron

element concentration of the residual solution after product separation, and the boron

element concentration can be obtained through the ICP experiment. Therefore, the

mass percentage of boron element in ZIF-67/RGO-B (wtB%) can be obtained through

the equations of (1) and (2).

S2

The specific ICP experiment is as follows: 0.5 ml of the residual solution after

product separation was placed in a 1000 ml beaker, and diluted with 550 ml of

deionized water, and then 5-10 ml of the diluted solution was placed in a sample tube

as the sample to be tested. The ICP experiment was conducted with an Optima 7300

DV spectrometer (Perkin-Elmer). The cooling flow, auxiliary flow, carrier gas and

sample flow during the experiment process were 15, 0.2, 0.8 L·min-1 and 1.5 ml·min-1,

respectively.

The total volume of the residual solution after product separation was about 100

ml, and the ICP result showed that the concentration of boron element in the diluted

solution was 2.657 mg·L-1. Therefore, the mass of boron element that remained in the

solution at the end of the reaction (m2) was about 0.292 g. Meanwhile, the mass of the

product ZIF-67/RGO-B (m) was 0.761 g and the mass of the boron element added

before the reaction start (m0) was 0.340 g (the mass of Na2B4O7·10H2O added before

the reaction start was 3 g). Thus, the mass percentage of boron element in

ZIF-67/RGO-B (wtB%) was about 6.3%.

TG results of as-prepared samples

Thermogravimetric Analysis (TGA) is made for studying the thermal stability of

the as-prepared samples. Fig. S1 shows the TGA curves of RGO, ZIF-67,

ZIF-67/RGO and ZIF-67/RGO-B under the condition of air atmosphere and the

heating rate of 20 °C·min-1. Obviously, RGO shows a continuous decomposition

curve, which is attributed mainly to the removal of physically adsorbed water and the

S3

continuous decomposition of the graphene skeleton. As can be seen from Fig. S1,

ZIF-67 begins to decompose at 340 °C, which is mainly attributable to the removal of

adsorbed water and the decomposition of imidazole ring in ZIF-67 [1]. In addition, it

is clear that the maximum degradation temperature of ZIF-67/RGO is decreased,

which is mainly because of the high thermal conductivity of RGO and the catalytic

degradation effect of ZIF-67. However, the char yield of ZIF-67/RGO at 700 °C is

higher than those of ZIF-67 and RGO, which is mainly because of the barrier effect of

RGO and the catalytic carbon effect of ZIF-67. Besides, the char yield of the sample

at 700 °C is further improved after borate ions are adsorbed onto the surface of ZIF-

67/RGO.

Fig. S1. TG curves of as-prepared samples

Thermal behavior of pure EP and EP composites

The glass transition temperature (Tg) is very important in the application of

S4

polymer materials and it is the macro embodiment of the transformation of polymer

movement. Normally, it is measured by using differential scanning calorimetry (DSC)

apparatus. Fig. S2 displays the Tg results of EP and EP composites. Obviously, the Tg

value of EP0 is 112.6 °C. However, compared with EP0, the Tg values of the EP

composites are improved in varying degrees, mainly because that the movement of

polymer segments is restricted to some extent by adding different flame retardants.

Fig. S2. DSC curves of neat EP and EP composites

Table S1 TG data of neat EP and EP composites

Samples T10% (°C) Tmax (°C) Char yield (%)

EP0 388.0 394.6 0.23

EP1 386.1 392.9 0.65

EP2 316.2 374.3 1.57

EP3 311.9 358.3 1.99

EP4 314.1 364.2 3.43

XPS results of char residue of pure EP and EP composites

XPS is also used for analyzing the char layers of the above EP composites

S5

obtained from the cone calorimeter test. Fig. S3 displays the C1s spectra of the char

layers of three samples (the specific values are displayed in Table S2). Obviously,

there are three peaks in the C1s spectra of three samples appearing at 284.7, 285.8 and

287.5 eV, respectively, corresponding to C-C bond, ether bond or C-O bond in

hydroxyl group and C=O bond in carbonyl group [2]. Generally, the thermal oxidation

resistance of materials can be estimated by calculating the ratio of Cox (representing

the total amount of oxidized carbons) with Ca (representing the amount of non-

oxidized carbons). In combination with Fig. S3 and Table S2, it is evident that the

Cox/Ca value of pure EP is 0.74. However, compared with EP0, the Cox/Ca values of

the composites with 2 wt% ZIF-67/RGO and ZIF-67/RGO-B are decreased to varying

extents, respectively. Specifically, the Cox/Ca value of EP4 is decreased the most

obviously, to 0.57. These results show that ZIF-67/RGO-B would enhance the

thermal-oxidation stability of the char layer more availably, indicating that

ZIF-67/RGO-B has a better effect on reducing the fire risk of polymer.

Fig. S3. C1s spectra of char residue of EP composites: (a) EP0 (b) EP3 (c) EP4

Table S2 Results of C1s XPS of char residue of EP composites

Sample C-C C-O C=O Cox/Ca

S6

area (%) area (%) area (%)

EP0 57.3 24.4 18.3 0.74

EP3 60.3 23.2 16.5 0.66

EP4 63.5 21.2 15.3 0.57

Fig. S4 (a) and (b) show the N1s spectra of the char layers of EP0 and EP4

obtained from the burning test (the specific values are displayed in Table S3). Fig. S4

displays that there are two peaks appearing at 400 and 398.3 eV, respectively,

corresponding to the binding energy of C=N and C-N. According to Fig. S4 and Table

S3, the content of C=N in the char layer of EP4 is increased from 62.1% to 69.3%,

and the content of C-N in the char layer of EP4 is decreased from 37.9% to 30.7%,

after the addition of 2 wt% ZIF-67/RGO-B. The formation of the conjugated structure

between carbon and nitrogen atom is due mainly to the combination between the

decomposition fragments during the combustion process [3]. The above results show

that the addition of ZIF-67/RGO-B could facilitate the generation of C=N in the char

layer. In the burning process of composites, ZIF-67/RGO-B could facilitate the

generation of a cross-linked structure and char layer, so that there are more

disintegration debris in the condensed phase to improve the stability of the char layer

and inhibit the release of toxic gas.

S7

Fig. S4. N1s spectra of char residue of EP0 (a) and EP4 (b)

Table S3 Results of N1s XPS of char residue of EP composites

Sample C=N (%) C-N (%)

EP0 62.1 37.9

EP4 69.3 30.7

S8

References

[1] E.H. Zhang, X. Yu, S.Q. Ci, J.C. Jia, Z.H. Wen, Porous Co3O4, hollow

nanododecahedra for nonenzymatic glucose biosensor and biofuel cell, Biosens.

Bioelectron. 81 (2016) 46-53.

[2] W.Z. Xu, B.L. Zhang, X.L. Wang, G.S. Wang, D. Ding, The flame retardancy and

smoke suppression effect of a hybrid containing CuMoO4 modified reduced graphene

oxide/layered double hydroxide on epoxy resin, J. Hazard. Mater. 343 (2017) 364-

375.

[3] X. Liu, J.Y. Wang, X.M. Yang, Y.L. Wang, J.W. Hao, Application of TG/FTIR

TG/MS and cone calorimetry to understand flame retardancy and catalytic charring

mechanism of boron phosphate in flame-retardant PUR-PIR foams, J. Therm. Anal.

Calorim. 1 (2017) 1-11.

S9