Chemical characterisation of polyoxybenzoate matrix composite materials by high performance liquid chromatography

Reagents and materials

p‐hydroxybenzoic acid (p‐HBA, 99·0%) was purchased from Tianjin Bodi Chemical Co., Ltd (Tianjin, China) and was used as standard substance after recrystallised three times in distilled water. MoS₂ colloidal powder (1–3 μm) and polyamide 66 powder (PA 66, 80–120 grids) were purchased from Shanghai Chemical Reagent Research Institute (Shanghai, China). Methanol (HPLC grade) and other analytical reagents were obtained from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). p‐acetoxybenzoic acid (p‐ABA) was prepared by refluxing p‐HBA with a 10 mol.‐% excess of acetic anhydride at 130–150°C. Zinc acetate was prepared by the reaction of ZnO and CH₃COOH and was recrystallised three times before use.

Experiments and instruments

Polyoxybenzoate based coatings were fabricated from the prescribed original materials (Table 1) on test blocks (1045 steel, 19×11×11 mm in size), in which 1 wt‐%Zn(OAc)₂ was added as a catalyst. First, the preparation of the composite materials was carried out at 160–170°C for 2 h in a 25 mL glass test tube. The mouth of the test tube was plugged with a block of cotton so as to inhibit the sublimation of p‐ABA. A little bag of alkali lime was wrapped in the cotton in an effort to absorb the byproduct CH₃COOH. Second, the test blocks were heated to 290°C and kept for 4 h. Finally, the coatings were cooled to the room temperature. Each step was under the protection of N₂ (99·9%).

OPEN IN VIEWER

Table 1

Compositions of original materials for various coatings

Sample no.	p‐ABA/g	PA 66/g	MoS2/g
0	0·1000	0·0000	0·0000
1	0·0800	0·0200	0·0000
2	0·0810	0·0180	0·0102
3	0·0732	0·0102	0·0200
4	0·0546	0·0114	0·0308

Polyoxybenzoate based composites were then scraped from the test blocks and transferred into 2 mL water in an 8 mL test tube. Few patches of KOH were added into the tube. The mixture was smashed using a glass stick until a paste‐like mixture was obtained. Then the test tube was kept at 90°C for 1 h in a thermostat.

Approximately 5 mL water was added into the obtained mixture in the test tube after it was taken out from the thermostat. Then the mixture was then filtered and only the solution was collected. Drops of H₂SO₄ and patches of KOH were alternatively added into the solution to adjust the pH value until 3–5. The solution was diluted quantitatively to 50·00 mL in a volumetric flask, 10·00 mL of which was taken out and diluted again accurately to 50·00 mL in another flask.

The high performance liquid chromatographic system consisted of two LC‐10AT VP pumps, a DGU‐12A degasser, a CTO‐10AS VP column oven, a 15 cm×t4·6 mm (i.d.) analytical column packed with 5 μm silica particles bonded with octadecylsilane, a SPD‐10A VP UV‐vis detector, a SCL‐10A VP controller and a Class‐VP data system. All these are from Shimadzu Corporation (Kyoto, Japan). The affiliated injector was Rheodyne 6‐port sampling valve (Model 7725i) with a 20 μL sample loop. The eluted components were monitored at the wavelengths of 210, 254 and 270 nm. A mixture of 1∶1 (v/v) CH₃OH.H₂O was used as eluant at a flowrate of 1·0 mL min⁻¹.

The peak area versus concentration calibration curves of p‐HBA at 210, 254 and 270 nm were measured by HPLC analysis using p‐HBA/CH₃OH standard solutions of 0·1, 0·3, 0·5, 0·7 and 1·0 mg L⁻¹. The confirmation of p‐HBA and the recovery test were carried out via a standard addition method. p‐HBA in each solution was quantitatively determined via an external standard method.

Structural characterisation of polyoxybenzoate

Polyoxybenzoate was prepared according to equation (1)

OPEN IN VIEWER

Figure 10

(1)

The chemical structure of the synthesised polyoxybenzoate was characterised by Fourier transform infrared spectrometry4 and X‐ray diffraction in Fig. 2. From Fig. 2, the characteristic peaks of the synthesised polyoxybenzoate were clearly observed at 2θ of 19·62, 20·92, 23·70 and 28·98. The sharp diffraction peaks of the obtained polyoxybenzoate demonstrates its high crystallinity (>90%), which is in agreement with that reported elsewhere.8

OPEN IN VIEWER

Figure 2

Spectrum of X‐ray diffraction analysis of synthesised polyoxybenzoate: sharp diffraction peaks demonstrate high crystallinity (>90%)

Optimisation of polyoxybenzoate degradation

Polyoxybenzoate was depolymerised in the KOH solution (of CH₃OH and H₂O for gas chromatography and HPLC respectively) corresponding to equation (2)

OPEN IN VIEWER

Figure 11

(2)

In Fig. 3, the peak area stands for the concentration of p‐HBA degraded from polyoxybenzoate in the KOH solution. Therefore, the curves in Fig. 3 represent the effect of time on polyoxybenzoate degradation efficiency. From Fig. 3, the degradation efficiency of polyoxybenzoate in both pure polyoxybenzoate and polyoxybenzoate based composites increased with the degrading time. When the degrading time was 40–50 min, the degradation efficiency reached the maximum. In order to ensure the complete degradation of polyoxybenzoate, the authors kept the smashed mixture of polyoxybenzoate or polyoxybenzoate based composites and KOH at 90°C in water bath for 60 min.

OPEN IN VIEWER

Figure 3

Effect of time on polyoxybenzoate degradation efficiency (90°C): rise of peak area represents increase in p‐HBA in solutions, i.e. increase in degradation efficiency of polyoxybenzoate

Determining polymerisation degree of polyoxybenzoate by high performance liquid chromatography

From equation (2), it seems that CH₃COOH, p‐HBA degraded from polyoxybenzoate and the molar ratio of p‐HBA/CH₃COOH can be used for the determination of polymerisation degree by a high performance liquid chromatographer. The maximum absorption wavelength λ_max was 210 and 254 nm for CH₃COOH and p‐HBA in UV spectra respectively. Considering that C₆H₅OH (λ_max ≈270 nm)3 might be used for the determination of polymerisation degree of the commercial polyoxybenzoate in the future, a standard solution of p‐HBA and acidified degradation products of composites was analysed at 210, 254 and 270 nm for comparison.

Figure 4 shows the calibration curves of p‐HBA, the formula, and the correlation coefficients by HPLC at 210, 254 and 270 nm. All linear correlation coefficients were greater than 0·99. Obviously, the sensitivity at 270 nm was much lower than that at 210 and 254 nm. Sensitivities and linear correlation coefficients obtained at 210 nm for each concentration were very close to that obtained at 254 nm. However, 210 nm is close to the UV cut‐off wavelength of CH₃OH (205 nm), which will result in a larger measurement error. Meanwhile, more impurity peaks and even an obvious negative peak appeared in HPLC chromatograms at 210 nm (Fig. 5). Figure 6 is the mass spectrum of the peak at 2·53 min in Fig. 5, in which the molecular ion peak 137·47 m z⁻¹ confirmed p‐HBA in the degraded products. In conclusion, 254 nm was the most suitable wavelength for determining the polymerisation degree of polyoxybenzoate by HPLC.

OPEN IN VIEWER

Figure 4

Working curves of peak area as functions of concentrations of p‐HBA by HPLC at 210, 254 and 270 nm

OPEN IN VIEWER

Figure 5

Comparison of HPLC chromatograms of sample no. 3 at 210 and 254 nm: there is negative peak in curve at 210 nm

OPEN IN VIEWER

Figure 6

Mass spectrum of p‐HBA from HPLC/MS analysis of degradation products of polyoxybenzoate based composites, peak 137·47 m z⁻¹ represents molecular ion peak of p‐HBA

It is infeasible to calculate the polymerisation degree from the CH₃COOH content in the acidified degradation products of polyoxybenzoate because detecting CH₃COOH is likely to result in a relatively large measuring error. The molar absorption coefficient ϵ of CH₃COOH by a UV detector was only experimentally 1/455 of that of p‐HBA. Meanwhile, CH₃COOH was failed to be detected sometimes, especially for polyoxybenzoate of a high polymerisation degree (e.g. >100). Similar problems would be encountered when the polymerisation degree was determined based on the molar ratio of p‐HBA to CH₃COOH. The polymerisation degree of polyoxybenzoate was thus determined by detecting p‐HBA degraded from polyoxybenzoate or polyoxybenzoate based composites.

Effect of polymerisation conditions on polymerisation degree of polyoxybenzoate

From Fig. 7, the polymerisation degree of polyoxybenzoate increased with the reaction time and temperature, especially when catalysed with Zn(OAc)₂. The maximum polymerisation degree at 290°C catalysed with Zn(OAc)₂ was 2·96 times that without Zn(OAc)₂ at the same temperature. While at 260°C, the polymerisation degree with Zn(OAc)₂ was 4·83 times that in the absence of Zn(OAc)₂. It is evident that the addition of Zn(OAc)₂ in reactants can effectively increase the polymerisation degree of polyoxybenzoate, in particular at a lower reaction temperature (e.g. 260°C). This may be because the addition of Zn(OAc)₂ can reduce the activation energy more remarkably at a lower temperature than that at a higher temperature. However, only polyoxybenzoate of a low polymerisation degree (e.g. <5) was obtained without Zn(OAc)₂ even at a higher temperature 290°C. Meanwhile, the polymerisation degree was still lower than 30 even for the polyoxybenzoate obtained at 290°C, catalysed with Zn(OAc)₂. Reasons may be as follows: the temperature gradient exists within the materials because the original materials were heated with no stirring during the fabrication of the composite coatings. However, CH₃COOH might be taken away by the N₂ flow, or absorbed by the alkali lime wrapped in the cotton; CH₃COOH will not be eliminated from the reaction system completely. The system cannot be vacuumed, because the p‐ABA should be prevented from sublimating out of the reactor.

OPEN IN VIEWER

Figure 7

Determined polymerisation degree of polyoxybenzoate by HPLC at varying reaction times and temperatures under catalysis or not

In terms of Carothers’ theory, the polymerisation reaction of polyoxybenzoate follows the second order rate kinetics under catalysis, and hence the polymerisation degree should rise linearly with the increase in reaction time.7 However, polymerisation degree increased exponentially with the reaction time. This may be that the curve of polymerisation degree versus time located at the non‐linear part before the linear one.

In the present research, moderate polymerisation degree of polyoxybenzoate is necessary for ensuring both an adequate load and a low friction coefficient.9 The synergistic effect of polyoxybenzoate and PA 66 and MoS₂ on the tribological properties of the lubrication coating can be accomplished only when the polymerisation degree of polyoxybenzoate is <39, as only the polyoxybenzoate of low polymerisation degree (e.g. <39) was observed to transform into liquid state under the shear stress during sliding. Meanwhile, bubble defects were found on the surface of the coating over 290°C, which will reduce significantly the coating's bonding strength and abrasion resistance. Hence, the optimised reaction temperature is 290°C in this research.

Composition of polyoxybenzoate based composite materials

Figure 8 demonstrates the results of recovery and reproducibility tests for various coatings via HPLC at 254 nm. From Fig. 8, relative standard deviations of the retention time and peak areas were <1%; the recoveries were 87·88–110·17% for four different composite materials and the pure polyoxybenzoate. The good reproducibilities and high recoveries mean that HPLC is a suitable approach for the determination of the polymerisation degree of polyoxybenzoate and the mass fraction of polyoxybenzoate in composites.

OPEN IN VIEWER

Figure 8

Results of recovery and reproducibility tests for various coatings via HPLC analysis at 254 nm

Figure 9 show that blocking test tube mouth with cotton helps to inhibit the sublimation of p‐ABA to some extent during the preparation of polyoxybenzoate based composite materials. The total mass of the coatings deviated 8·3–35% from the theoretical values when blocked with cotton, while the deviation was 22–57% without blocking.

OPEN IN VIEWER

Figure 9

Effect of blocking on total mass of prepared polyoxybenzoate based composite coatings

Table 2 shows the mass fraction of polyoxybenzoate in various composites determined using HPLC at 254 nm. In Table 2, deviation of the mass fraction in polyoxybenzoate based composites were observed to be 12·68–20·21% when blocked and 34·97–49·18% without blocking from the theoretical values, which is in agreement with the deviation of total mass of the composites. This shows that the technology for obtaining polyoxybenzoate based composites of accurate composition is not yet mature. The possible reason is that the sublimation of p‐ABA over 180°C cannot be inhibited thoroughly though certain measures have been taken.

OPEN IN VIEWER

Table 2

Effect of blocking on mass fraction of polyoxybenzoate in several polyoxybenzoate based composites

No.	Polyoxybenzoate (theoretical)/%	Blocked		Unblocked
		Polyoxybenzoate (measured)/%	Deviation/%	Polyoxybenzoate (measured)/%	Deviation/%
1	72·83	58·11	−20·21	37·01	−49·18
2	65·82	55·29	−15·99	36·12	−45·12
3	61·90	52·76	−14·76	35·38	−42·84
4	43·20	37·72	−12·68	28·09	−34·97

It can be observed that the deviation of the total mass of coatings and mass fraction of polyoxybenzoate reduced with the increase in the mass fraction of MoS₂ in the original materials either blocked or not. It may be that nano‐MoS₂ powder is of help to reduce the sublimation of p‐ABA by fixing it in the mixture so that the total mass of coatings and mass fraction of polyoxybenzoate approach the theoretical values.

Eliminating the byproduct, CH₃COOH, thoroughly helps increase the degree of polymerisation of polyoxybenzoate. On the other hand, sealing the reactor is beneficial to lowering the sublimation of p‐ABA so as to reduce the deviation of total mass of composite materials or mass fraction of polyoxybenzoate in the composites. There is a conflict between them. Measures beneficial for the elimination of CH₃COOH will usually lead to a faster sublimation of p‐ABA. For example, vacuuming the reactor can eliminate CH₃COOH more effectively; however, it will lead to a faster sublimation of p‐ABA in original materials. Fabricating the composite coatings in a hermetically sealed reactor may prevent p‐ABA from sublimating; nevertheless, it will run the risk of explosion in the experiment. Currently, no better measures are beneficial for both eliminating CH₃COOH and lowering the sublimation of p‐ABA.