Solid-phase microextraction (SPME), a relatively new sample preparation technique introduced by Arthur and Pawliszyn, is a simple and effective technique which integrates sampling, extraction, and concentration (Arthur and Pawliszyn 1990; Zhang and Pawliszyn 1993; Pawliszyn 1997). Owing to the advantages of solvent-free nature, SPME has been widely used in the analysis of environment, food, biological forensic and so on (Vuckovic et al. 2011; Ouyang et al. 2011). The sorbent coated on the SPME fiber is the key factor affecting the sensitivity and selectivity of the analytes. However, the availability of commercial SPME fibers such as PDMS, PDMS/DVB, CAR/PDMS, PA are quite limited. And the silica-based fibers also have weak mechanical strength and relatively low thermal instability (usually 240–280°C) (Derouiche et al. 2007; Yang et al. 1998; Llompart et al. 1999; Landin et al. 2001). To overcome these problems, the exploitation of novel sorbents for SPME fibers is still challenging work.

Metal–organic frameworks (MOFs) are a novel class of porous materials built from metal ions and organic ligands which are extended infinitely into one, two or three dimensions mainly through the metal–ligand coordination, hydrogen bonding and π–π stacking. MOFs have various potential applications due to its large surfaces, modifiable pores and high thermal stability (Lin et al. 2002; Lee et al. 2009). MOFs have also been successfully explored as sorbents for solid-phase extraction (SPE) (Aquino et al. 2010; Barreto et al. 2010; Carvalho et al. 2009; Gu et al. 2010, 2012; Wang et al. 2012, 2014; Yang and Yan 2013; Zhou et al. 2006), solid-phase microextraction (SPME) (Chang et al. 2011; Cui et al. 2009; Gu et al. 2010; Wu et al. 2014; Yang et al. 2012), and as stationary phases of the chromatography column (Chang et al. 2010; Liu et al. 2012; Yang et al. 2011). To date, a few porous MOFs, such as MIL-101(Cr) (Xie et al. 2015; Zang et al. 2013), MOF-199 (Cui et al. 2009), ZIF-8 (Chang et al. 2011), MIL-53(Al) (Chen et al. 2012), MAF-X8 (He et al. 2013), ZIF-90 (Yu and Yan 2013), MIL-88B (Wu et al. 2014), UiO-66 (Shang et al. 2014), MOF-177 (Wang et al. 2015) have been all reported as SPME fibers.

Herein a porous MOF, MOF-177, possessing high surface area (4500 m2 g−1, BET) and large pore windows, was also explored as SPME adsorbent (Wang et al. 2015). Briefly, this work can be regarded as a continuation of developing the method for other classes of compounds. The environmental phenols were selected as analytic targets for their trace level, high toxicity, widespread environmental occurrence and resistance to biodegradation (Chae et al. 2004). Totally five kinds of phenols with different polarity (the log KOW ranges from 3.49 to 8.18) and molecular sizes (from 6.57 to 7.52 Å) were investigated as analytes from environmental water. The HS-SPME method was employed before the GC/MS analysis, and three kinds of real water samples (river water, lake water and waster water) were also determined.

Materials and Methods

The materials, chemical reagents (except the standard) and the equipment employed in this work were the same as in the previous work (Wang et al. 2015). The standard of 2-methylol-phenol (2-MP), 4-methylolphenol (4-MP), 2,4-dimethylphenol (2,4-DMP), 2,4-dichlorphenol (2,4-DCP), and 3-methyl-4-chlorophenol (3-M-4-CP) were purchased from Sigma-Aldrich (Saint Louis, USA). The stock solution of the mixed standard were prepared as 10 mg mL−1 in methanol, and stored at 4°C in darkness. The water samples were taken from different local areas, South of China. Before the SPME experiment, all samples were filtered with the 0.45 μm filter membranes and stored in precleaned brown glass bottles.

For the analysis, the column temperature of GC/MS was programmed as follows: 50°C for 2 min, then heated to 130°C at 5°C min−1, and finally to 280°C at 25°C min−1, with a hold time of 6 min. The injector port was maintained at 250°C and the SPME fiber was injected in splitless mode. After 5 min of desorption time, the data were acquired and processed using Agilent GC Solution software. The interface and ionization source temperature were 280 and 230°C, respectively. The analysis was performed in SIM mode, and the molecular ion (m/z 108, m/z 107, m/z 122, m/z 162, m/z 142) was used as the target ion for detection (ionization mode: electron ionization with 70 eV energy).

The fabrication of the SPME fiber was slightly modified on the basis of the reported work (Wang et al. 2015). Briefly, the stainless steel wire of a 5 μL GC micro-syringe was etched with hydrofluoric acid (38%, w/w) for 15 h. The MOF-177 powder was attached on the wire through silicone sealant diluted with n-hexane (w/v: 200 mg/1 mL). The coatings obtained after two coating cycles were around 30 μm thick and 1 cm long. Finally, the redundant coatings were removed and aged for 10 min at 250°C under the GC inlet. All HS-SPME extractions were carried out according to the publicated method (Wang et al. 2015).

Results and Discussion

The characterization of MOF-177 (X-ray powder diffraction and SEM, Fig. 1) were the same as our previous work (Wang et al. 2015). To obtain the optimization extraction efficiency, various experimental parameters influencing the extraction progress were investigated, including extraction temperature and time, ionic strength and pH value. To simplify the experimental step, the desorption time and temperature were set at 5 min and 250°C, which can assure the analytes be desorbed completely in this experiment. And the agitation speed was set at 700 rpm.

Fig. 1
figure 1

The XRD (a) and SEM (b) characterization of the coated fiber

Full size image

The effect of extraction temperature was studied from 30 to 70°C. As seen from Fig. 2a, the peak area increase steadily with the increase of temperature before 50°C. After 50°C, the extraction efficiency of 2,4-DCP and 3-M-4-CP still increase, however, for 2,4-DMP, 2-MP and 4-MP, the efficiency occur slight decline tendency. High extraction temperature is in favor of the release of analysts to the upper gaseous phase, however, too high extraction temperature is unfavorable for the exothermic adsorption of the analytes onto the SPME fiber (Wu et al. 2014). 60°C was chosen as the final extraction temperature.

Fig. 2
figure 2

The effect of experimental parameters on extraction efficiency. a Extraction temperature; b extraction time; c the NaCl addition; d the pH value. (Other conditions: concentration, 10 μg L−1; desorption temperature, 250°C; desorption time, 5 min; stirring rate, 700 rpm)

Full size image

The extraction time was also an important factor to influence the final result. The extraction efficiency always increases before the vapor–liquid phase reaches equilibrium. Thirty to seventy minutes were tested individually in the extraction process. As seen from Fig. 2b, the peak area increases with the extending of time. After 60 min, the peak area occur slight increase or decline tendency. The final extraction time was set at 60 min.

The effect of ionic strength on the peak area of the phenols was investigated in the NaCl addition range of 1–2.5 g with the sample volume of 10 mL. As presented in Fig. 2c, the peak area increase dramatically when the NaCl addition is lower than 1.5 g, and appear decrease with the further addition of NaCl. This may be ascribed that the salt-out effect plays the main role in the HS-SPME process, however, the excessive Na+ may also weaken the π–π interaction between the ligands and analytes. Therefore, further experiments were performed at the salt addition of 1.5 g.

The effect of pH value was also investigated from 3 to 9. The peak area increase when the pH value was improved from 3 to 5, and decrease with the further rise of pH value (Fig. 2d). When the solution was at lower pH, the release of HCl molecules may compete with the analytes adsorbed on the SPME fiber. If the solution appears basic, the solubility of phenols increase dramatically which cause the volatile amount of phenols decrease. For the next experiment, the pH value was adjusted to 5.

The extraction performance of MOF-177-coated fiber was compared with 100 μm PDMS, 65 μm PDMS/DVB, 85 μm PA, the etched stainless fiber and the adhesive-coated fiber, respectively. As shown in Fig. 3a, the MOF-177-coated fiber shows the best extraction efficiency towards the target analytes. For the analytes, the value of log KOW are followed an increasing order (4-MP < 2-MP < 2,4-DMP < 3-M-4-CP < 2,4-DCP). The minimum dynamic diameters of target molecules range from 3.9 Å (4-MP) to 5.4 Å (3-M-4-CP). The extraction performance gives an increasing order of 2-MP < 4-MP < 3-M-4-CP < 2,4-DMP < 2,4-DCP. The hydrophobicity, molecular size of phenols and the cavity characteristic of MOF-177 play the main roles towards the extraction efficiency. For MOF-177, the narrowest edge of pores is 10.8 Å, which allows all the analytes to enter into the pores. The H3BTB ligand contains four phenyl rings, which can afford abundant π–π and C–H…π interactions to phenols.

Fig. 3
figure 3

a Comparison of extraction efficiency of MOF-177-coated fiber, PDMS, PDMS/DVB and PA fiber; b the tic chromatogram of surface water sample (below) and the spiked surface water sample (up). Conditions: spike level, 10 μg L−1 for each analyte; sample volume, 10 mL; extraction temperature, 60°C; extraction time, 60 min; agitation speed, 700 rpm; NaCl addition, 1.5 g; pH value, 5

Full size image

The limits of detection (LOD), correlation coefficients, repeatability were determined under the optimized conditions for phenols analysis (Table 1). The linearity range of the method was measured from 0.1 to 50 μg L−1. The correlation coefficients (R2) were found to be between 0.990 and 0.997. The LODs (S/N = 3) were in the range of 0.015–0.043 μg L−1. The LOQs (S/N = 10) were in the range of 0.05–0.144 μg L−1. The repeatability rates obtained from six replicate samples were from 2.8% to 5.5% and the reproducibility rates from the three fibers were from 4.2% to 7.4%. Among the direct HS-SPME or derivatization-HS-SPME method established for phenols, the LODs obtained in this work were lower or close to those reported results (Table 2) (Anbia et al. 2012; Regueiroa et al. 2009; Abolghasemi et al. 2016; Peñalver et al. 2002).

Table 1 Analytical performance of the coated fiber for the SPME of phenols under the optimized conditions
Full size table
Table 2 Comparison of different fibers for the SPME of phenols
Full size table

The established method was also applied to three kinds of real water samples (surface water, river water and waste water). The results were given in Table 3. For the surface water and river water samples, only 2,4-DCP was found in surface water2 and river water4 (Gao et al. 2008). Four kinds of analytes except 3-M-4-CP were found in waste water samples, ranging from 0.26 to 9.78 μg L−1, and the recoveries were between 84.5% and 98.6% for phenols (spiked at 10 μg L−1).

Table 3 Analytical results for the determination of phenols in real water samples
Full size table

In this work, we have fabricated a thermal stable MOF-177-coated SPME fiber with an adhesive method, which offered good extraction efficiency for phenols than commercial fibers. The established method has good repeatability and relative LODs compared with the derivation method. The good stability and long lifetime of the prepared fiber suggest that the MOF-177-coated SPME fiber is promising in the analysis of real samples in separation field.