High-efficiency cucurbit[7]uril capillary column for gas chromatographic separations of structural and positional isomers

Abstract

Here we report the first example of cucurbit[7]uril (CB7) diluted in trifluoropropylmethylpolysiloxane (OV210) as the binary stationary phase (CB7–OV) for high-resolution gas chromatographic (GC) separations. The as-fabricated CB7–OV capillary column achieved extremely high column efficiency of 4660 plates per m. This is the highest column efficiency ever reported for cucurbit[n]urils for GC separations. Most importantly, the CB7–OV column exhibited superior selectivity and resolving ability for structural and positional isomers covering apolar to polar analytes, showing advantages over the neat CB7 and OV210 columns. Moreover, it showed good column repeatability and reproducibility with RSD values of 0.01–0.09% for run-to-run and 0.03–0.47% for day-to-day, and 2.9–5.2% for column-to-column, respectively, and column thermal stability up to 260 °C. The established highly-efficient strategy promotes taking full advantage of cucurbit[n]urils as GC stationary phases and may also be feasible for other advanced materials in separation science.

---

1 Introduction

Cucurbit[n]urils (CBn, n = 5–8, 10) are macrocyclic compounds comprising n glycoluril units shaped like a pumpkin with one hydrophobic cavity and two identical polar carbonyllaced portals and have high thermal stability.^1,2 For example, CB7 is thermally stable up to 370 °C. Their unique amphiphilic structures offer high molecular recognition capability for compounds of diverse types.³ Specifically, the apolar interior provides an interaction site for nonpolar to low polar analytes while the polar carbonyl-laced portals offer high affinity for moderate polar to polar analytes. The methine/methene protons can also contribute to molecular recognition by selectively interacting with H-bonding analytes. They have attracted growing attention in supramolecular self-assembly,^4–6 coordination chemistry⁷ and others,⁸ which usually perform in aqueous solution in the presence of inorganic salts due to the poor solubility of CBn solids in organic solvents. In solid states, CBn molecules tend to form supramolecular aggregates by noncovalent bonding to each other.² As such, their recognition ability in solid aggregates for guest molecules would be harmed due to the limited accessible recognition sites by guest molecules.

Recently, advancements have been made for utilizing CBn as stationary phases in gas chromatography (GC)^9–13 and liquid chromatography.¹⁴ They show selective GC separations and high column stability. However, fabricating high-efficiency columns has been the major challenge for CBn stationary phases due to their poor solubility in organic solvents used for column fabrication. In our previous work, we made great improvements by using sol–gel coating method^11,12 and by static coating method with the aid of a guanidinium-based ionic liquid.¹³ As proven, the ionic liquid was highly efficient in achieving high column efficiency but it is not easily available and costs high for its synthesis. Therefore, exploring other efficient materials would be favorable for the practical use of high-efficiency CBn columns.

Polysiloxanes are a large family of substituted polymers with various functional groups of different polarity and have been widely used in GC separations. Due to their good film-forming ability, a few of them, such as OV1701 in most cases, were also used as a diluent in the coating solution of different separation materials to improve their column efficiency and selectivity as well.^15–20 The related reports inspired us that they might also be workable for CBn stationary phases. The polysiloxane chosen for this work was supposed to be able to enhance the solubility of CBn solids in dichloromethane by disorganizing the supramolecular aggregates of CBn molecules through H-bonding interaction with the carbonyl groups or methine/methene protons of CBn molecules. Most importantly, the disorganization of aggregates may have more recognition sites accessible for guest molecules and also favor the formation of uniform coating on the inner capillary column, which may simultaneously improve the selectivity and column efficiency of the fabricated column. Taking these into account, trifluoropropylmethylsiloxane (OV210) seemed the suitable one for the task, which has not been reported for the purpose.

In this study, we report the first example of CB7 diluted in OV210 as the binary stationary phase (CB7–OV) for high-resolution GC separations. After the column efficiency and McReynolds constants were determined, CB7–OV column was investigated for its selectivity and retention behaviours by fourteen groups of structural or positional isomers and two mixtures covering diverse types of analytes. Comparison was also made with three reference columns, two neat CB7 and OV210 columns and one commercial DB-35 column having quite close polarity to CB7–OV column. Furthermore, column repeatability, reproducibility and thermal stability were examined. This work demonstrates the advantages of the proposed strategy in addressing the challenging problem of CBn stationary phases for GC separations.

2 Experimental

2.1 Materials and equipment

All the chemicals in this work were of at least analytical grade and dissolved in dichloromethane. CB7 stationary phase was synthesized by the method described in ref. 9. 50% trifluoropropylmethylpolysiloxane was purchased from Beijing Chemical Reagent Company (Beijing, China). Untreated fused-silica capillary tubing (0.25 mm i.d.) was purchased from Yongnian Ruifeng Chromatogram Apparatus Co., Ltd. (Hebei, China). A DB-35 capillary column containing 35% phenylmethyl polysiloxane (10 m × 0.25 mm, i.d.) was purchased from Agilent Technologies.

An Agilent 7890A gas chromatograph equipped with a split/splitless injector, a flame ionization detector (FID) and an autosampler was used for GC separations. All the separations were performed under the following GC conditions: nitrogen of high purity (99.999%) as carrier gas at a flow rate of 1 mL min⁻¹, injection volume of 0.1 μL, injection port at 250 °C, split injection mode at a split ratio of 50 : 1 and FID at 300 °C.

2.2 Fabrication of the capillary columns of CB7–OV, CB7 and OV210

Prior to coating, a capillary column (10 m × 0.25 mm, i.d.) was rinsed with dichloromethane and dried with a nitrogen purge at 120 °C for 2 h, and then pretreated with the saturated solution of sodium chloride in methanol for the surface roughening of the capillary column. Then, the column was statically coated with the solution of CB7 and OV210 (2 : 1, w/w) in dichloromethane (CB7, 2.5 mg mL⁻¹) under 40 °C. During the coating process, one end of the capillary column was sealed when the solution was pumped into the pretreated capillary column and the other end was connected to a vacuum system to gradually remove the solvent under vacuum. The fabricated column was denoted as CB7–OV column. Similarly, two reference columns coated with only CB7 or OV210 were also fabricated and labeled as CB7 column and OV210 column, respectively. Prior to use, the columns were conditioned from 40 °C to 200 °C at 1 °C min⁻¹ and held at the final temperature for 7 h under a constant flow of nitrogen.

3 Results and discussion

3.1 Column efficiency and McReynolds constants

For the fabricated CB7–OV column, the Golay plot relating the height equivalent to a theoretical plate (HETP) with the flow rate of carrier gas was determined by n-dodecane at 120 °C. As shown in Fig. 1 and Table 1, the minimum HETP was 0.214 mm corresponding to the column efficiency of 4660 plates per m, which is the highest efficiency ever reported for CBn GC columns, as summarized in Table 2. This achievement can be attributed to the uniform coating of CB7–OV stationary phase on the capillary column. Additionally, the coating efficiency was obtained as 81.9%, calculated by the ratio of the theoretical minimum plate height to the measured minimum plate height.²¹

Fig. 1 Golay plot of CB7–OV column determined by n-dodecane at 120 °C.

Table 1 The coefficients of the Golay equation and the column efficiency for CB7–OV capillary column determined by n-dodecane at 120 °C

Parameters	n-Dodecane
Hmin (mm)	0.214
B (±sd)	0.042 (±0.001)
C (±sd)	0.288 (±0.002)
Column efficiency (plates per m)	4661
Flow rate (mL min^−1)	0.45

---

Table 2 Comparison of the column efficiency of CB7 columns fabricated by different methods

Column	Coating method	Column efficiency (plates per m)			Ref.
		Temperature	Analyte	n
CB7–OV	Static	120 °C	n-Dodecane	4661	This work
CB7	Static	100 °C	n-Dodecane	1400	9
CB7–IL	Sol–gel	100 °C	Naphthalene	2078	12
CB7–IL	Static	120 °C	n-Dodecane	4360	13

---

McReynolds constants are commonly used to evaluate the general or average polarity of a given stationary phase as the reference for column selection and retention prediction in practice. They can be determined by the five probe compounds, namely benzene (X′), 1-butanol (Y′), 2-pentanone (Z′), pyridine (S′) and 1-nitropropane (U′). In this work, the McReynolds constants of CB7–OV column as well as the reference columns were determined at 120 °C. Table 3 shows the moderate polarity of CB7–OV column, situating in-between the two neat phases. Besides, CB7–OV column had different elution sequence of the probes from the neat ones, indicating their different retention behaviors. Especially, 1-butanol retained longer on CB7–OV column than on CB7 column, suggesting the stronger H-bonding interaction of CB7–OV stationary phase with the proton donor possibly due to the synergistic effect of the binary stationary phase. Afterwards, the selectivity and retention behaviours of CB7–OV column were further explored as follows.

Table 3 McReynolds constants of CB7–OV capillary column^a

Stationary phase	X′	Y′	Z′	U′	S′	General polarity	Average
a X′, benzene; Y′, 1-butanol; Z′, 2-pentanone; U′, 1-nitropropane; S′, pyridine. Temperature: 120 °C.
CB7–OV	67	177	168	256	161	712	166
CB7	11	146	78	124	82	441	88
OV-210	184	264	283	378	242	1351	270
DB-35	102	142	145	219	178	786	157

---

3.2 Separation performance

Separation performance of CB7–OV column regarding selectivity, resolving capability and retention mechanism was investigated and evaluated by one mixture of 14 analytes of diverse types, fourteen groups of structural or positional isomers, and one more complex mixture of 26 aromatic compounds. Meanwhile, three reference columns, namely CB7, OV210 and DB35 capillary columns, were also employed for comparison so as to have a better understanding of the separation performance of the binary stationary phase.

Fig. 2 shows the separation results for the mixture of 14 analytes on CB7–OV column and two neat columns. As shown, CB7–OV column achieved baseline resolution of all the analytes in contrast to the coelution of some analytes on CB7 column (peaks 2–3, 9–10) and OV210 column (peaks 1–2 and 6–7). Particularly, the complete resolution of the first three analytes of heptanal, decane and benzaldehyde on the CB7–OV column suggested its advantageous resolving feature. In addition, it showed different retention behaviours from the neat ones, typically for the analytes of m-dibromobenzene, ethyl benzoate and 1,2,3-trichlorobenzene (peaks 7–9). CB7–OV column reversed the peaks 7–8 and peaks 8–9 in contrast to CB7 and OV210 column, respectively. Reasonably, the later elution of ethyl benzoate on CB7–OV column than CB7 column can be ascribed to the enhanced H-bonding interactions of its ester carbonyl group with the adjacent methane protons to the trifluoropropyl groups in OV210 and with the methane/methane protons of CB7 molecules.

Fig. 2 Separations of the mixture of 14 analytes on CB7–OV, CB7 and OV210 columns, respectively. Peaks: (1) heptanal, (2) n-decane, (3) benzaldehyde, (4) n-undecane, (5) 1-bromooctane, (6) nitrobenzene, (7) m-dibromobenzene, (8) ethyl benzoate, (9) 1,2,3-trichlorobenzene, (10) m-nitrotoluene, (11) o-chloronitrobenzene, (12) m-bromonitrobenzene, (13) n-pentadecane and (14) methyl dodecanoate. Temperature program: 50 °C for 1 min to 160 °C at 10 °C min⁻¹.

Moreover, a general trend can be found that CB7–OV column showed stronger retention for the alkanes (peaks 2, 4, 13) and the substituted benzenes with electron-withdrawing groups (peaks 6, 7, 9). The possible reason behind this finding can be elucidated as follows. As known, CB7 molecules in solid state tend to aggregate together through intermolecular H-binding of carbonyl groups with the methine/methene protons. Incorporation of CB7 into the polysiloxane may free some of the CB7 carbonyl groups originally H-bonded since the fluorine atoms of trifluoropropyl groups may form H-binding with the methine protons instead due to the much larger electronegativity of a fluorine atom than a carbonyl oxygen atom. As a result, more free carbonyl groups in the binary phase are available and accessible to specific analytes. Accordingly, the alkanes and electron-deficient aromatics may have more chance to interact with the nonpolar cavity via dispersion or with the electron-rich portals via π–π EDA and C–H⋯π interactions, respectively. Another notable advantage of CB7–OV column lies in its much improved peak shapes for polar analytes in contrast to CB7 column.

The above findings on the unique retention behaviour of CB7–OV column for alkanes and substituted benzenes inspired us to further explore its selectivity for their isomers. As a result, more than a dozen groups of isomer mixtures were examined, as shown in Fig. 3 and Table 4, respectively. Fig. 3 shows the separations of the isomer mixtures of hexanes, bromonitro benzenes and dimethylnaphthalenes on the CB7–OV and CB7 columns, respectively. It was interesting to note that CB7–OV column exhibited greatly enhanced resolving ability and prolonged retention for the three groups of the analytes although it shared the same elution order with CB7 column. For the hexane isomers, higher resolution was obtained by CB7–OV column than CB7 column that only had the resolution of 0.60 for the peaks 1–2 and 0.99 for peaks 2–3. In the binary phase, the linear hexane retained longer and baseline resolved with the adjacent analyte because it is more accessible to the cavity due to its smaller dynamic diameter (4.3 Å) than that of the CB7 portals (5.4 Å). With the similar reason, CB7–OV column achieved higher resolution for 2,2-dimethylbutane and 2-methylpentane.

Fig. 3 Separations of the indicated isomers on CB7–OV and CB7 columns, respectively. Temperature program for hexanes, 35 °C, at 0.5 mL min⁻¹; bromonitrobenzenes, 80 °C for 5 min to 160 °C at 10 °C min⁻¹; dimethylnaphthalenes, 100 °C.

Table 4 Parameters for the separations of the indicated isomers on CB7–OV column^a

Analytes	Elution order	tR	k	α	R
a The parameters tR, α, k and R refer to retention time, selectivity factor, retention factor and resolution, respectively.
Butanol	tert-	1.07	0.96	—	—
	sec-	1.20	1.20	1.25	3.81
	n-	1.27	1.34	1.11	1.72
	iso-	1.44	1.64	1.23	3.10
Pentanol	2-Methyl-2-butanol	1.20	1.24	—	—
	2-	1.41	1.63	1.31	4.29
	1-	1.90	2.55	1.56	6.64
Dimethylaniline	2,6-	10.34	18.25	—	—
	2,5-	10.49	18.54	1.02	1.84
	3,5-	10.70	18.93	1.02	2.53
	3,4-	11.03	19.54	1.03	3.80
PAHs	Phenanthrene	11.29	22.09	—	—
	Anthracene	11.42	22.35	1.01	1.44
Methylnaphthalene	2-	4.27	6.66	—	—
	1-	4.43	6.95	1.04	2.50
Dichlorobenzene	m-	12.67	11.59	—	—
	p-	13.40	12.31	1.06	2.16
	o-	15.77	14.67	1.19	7.52
Trichlorobenzene	1,3,5-	2.49	4.10	—	—
	1,2,4-	2.90	4.93	1.20	7.52
	1,2,3-	3.26	5.66	1.15	6.25
Trimethylbenzene	1,3,5-	4.93	3.90	—	—
	1,2,4-	5.27	4.23	1.09	4.98
	1,2,3-	5.70	4.67	1.10	6.54
Butylbenzene	tert-	5.13	4.10	—	—
	iso-	5.32	4.28	1.05	2.58
	n-	6.00	4.96	1.16	9.95
Nitrotoluene	o-	5.93	10.32	—	—
	m-	6.52	11.45	1.11	7.89
	p-	6.81	11.99	1.05	3.75
Chloronitrobenzene	m-	7.78	13.49	—	—
	p-	7.94	13.78	1.02	2.03
	o-	8.17	14.21	1.03	2.91

---

For the bromonitrobenzene isomers, CB7–OV column baseline resolved the three isomers (R = 1.78 for peaks 1/2, R = 3.12 for peaks 2/3), which were partially overlapped or coeluted on the other column. Complete separation of p-bromonitrobenzene from o-bromonitrobenzene can be attributed to the stronger π–π EDA interaction of p-bromonitrobenzene with the binary stationary phase owing to its more electron-deficient benzene ring than o-bromonitrobenzene. In addition, the much prolonged retention of the binary phase for these aromatics evidenced its preferential retention for the electron-deficient aromatics. This conclusion was also confirmed by other halogenated and nitrobenzene isomers, such as dichlorobenzenes, trichlorobenzenes, nitrotoluenes and chloronitrobenzenes, which are listed in Table 4 for brevity.

Moreover, CB7–OV column achieved high resolution of 1,4-dimethylnaphthalene and 2,3-dimethylnaphthalene probably through π–π and C–H⋯π interactions with the analytes of larger π system. Besides, it can also well separated the critical pairs of phenanthrene/anthracene and 1-methylnaphthalene/2-methylnaphthalene (Table 4). Table 4 also lists other isomers well separated on CB7–OV column, including alcohols, anilines and alkyl benzenes. Noteworthy, the binary phase also had high distinguishing capability for the H-bonding donors such as alcohols and anilines via H-bonding interactions.

Furthermore, we separated a more complex mixture of aromatic analytes covering arenes, halides, ethers, esters, aldehydes, ketones, anilines, nitro and nitriles on CB7–OV column in comparison to the commercial DB-35 column. As can be seen from Fig. 4, CB7–OV column achieved good separation for almost all the analytes, especially for phenetole, m-dichlorobenzene and benzaldehyde (peaks 9–11). In contrast, the latter two analytes totally overlapped on the other column. Their baseline resolution on the CB7–OV column may stem from their specific π–π EDA interactions with the binary phase. Additionally, the reversal elution of p-methylbromobenzene and benzonitrile (peaks 12–13) evidenced the different retention behaviour of the binary phase from the conventional phase despite their similar polarity. Moreover, CB7–OV column eluted ethyl benzoate (peak 18) behind m-dibromobenzene (peak 17) against their boiling points, possibly due to the comprehensive result of the H-bonding interaction from the ester carbonyl group with the adjacent methane protons to the trifluoropropyl groups of the polysiloxane, and the π–π EDA interaction of the electron-deficient benzene ring of the analyte with the electron-rich portals of CB7 molecules in the binary phase. In addition, the elution of 3,5-dimethylaniline (peak 20) behind 1,2,3-trichlorobenzene (peak 19) evidenced the contribution of H-bonding interactions to the retention of the binary phase. By contrast, DB-35 column eluted the analytes by their boiling points.

Fig. 4 Separations of the mixture of 26 aromatic analytes on CB7–OV and DB-35 columns, respectively. Peaks: (1) toluene, (2) chlorobenzene, (3) m-dimethylbenzene, (4) o-dimethylbenzene, (5) bromobenzene, (6) p-chlorotoluene, (7) tert-butylbenzene, (8) 1,2,4-trimethylbenzene, (9) phenetole, (10) m-dichlorobenzene, (11) benzaldehyde, (12) p-methylbromobenzene, (13) benzonitrile, (14) acetophenone, (15) nitrobenzene, (16) 2-chlorobenzenamine, (17) m-dibromobenzene, (18) ethyl benzoate, (19) 1,2,3-trichlorobenzene, (20) 3,5-dimethylaniline, (21) 3,4-dimethylaniline, (22) m-nitrotoluene, (23) m-chloronitrobenzene, (24) o-chloronitrobenzene, (25) m-bromonitrobenzene and (26) o-bromonitrobenzene. Temperature program: 40 °C for 1 min to 160 °C at 10 °C min⁻¹.

3.3 Column repeatability and thermal stability

Column repeatability and reproducibility were evaluated by the isomer mixtures of nitrotoluenes, chloronitrobenzenes, pentanols, butanols and trimethylbenzenes, respectively. As summarized in Table 5, the RSD values on the retention times were below 0.09% for run-to-run, less than 0.47% for day-to-day and in the range of 2.9–5.2% for column-to-column, respectively, demonstrating its good column repeatability and reproducibility. Also, its thermal stability was investigated by separation of the indicated isomer mixtures after the column was conditioned up to each of the temperatures of 200, 220, 240, 250 and 260 °C and held at each temperature for 2 h, respectively. As presented in Fig. 5, the column still maintained high selectivity for the isomers after the several cycles of conditioning up to 260 °C, showing its good column thermal stability.

Table 5 Repeatability and reproducibility of CB7–OV column on retention times (min) for separation of the indicated isomer mixtures

Analyte	Elution order	Run-to-run (n = 4)		Day-to-day (n = 4)		Column-to-column (n = 4)
		Mean	RSD (%)	Mean	RSD (%)	Mean	RSD (%)
Nitrotoluene	o-	5.931	0.02	5.930	0.04	6.023	4.0
	m-	6.524	0.01	6.523	0.03	6.633	3.8
	p-	6.806	0.01	6.805	0.03	6.922	3.6
Chloronitrobenzene	m-	7.779	0.04	7.779	0.06	7.879	3.4
	p-	7.936	0.04	7.937	0.05	8.042	3.4
	o-	8.164	0.05	8.166	0.05	8.287	2.9
Pentanol	2-Methyl-2-butanol	1.202	0.03	1.203	0.13	1.252	3.4
	2-	1.411	0.05	1.412	0.23	1.474	4.1
	1-	1.903	0.09	1.908	0.47	2.030	5.2
Butanol	tert-	5.132	0.02	5.135	0.08	5.136	4.5
	sec-	5.318	0.03	5.323	0.10	5.317	4.4
	n-	5.996	0.03	6.003	0.12	6.005	4.0
Trimethylbenzene	1,3,5-	4.933	0.02	4.928	0.07	4.939	4.8
	1,2,4-	5.268	0.01	5.264	0.07	5.276	4.5
	1,2,3-	5.702	0.01	5.704	0.08	5.736	4.2

---

Fig. 5 Effect of column conditioning temperatures on selectivity factor (α) on CB7–OV column evaluated by separations of four mixtures of positional isomers. Mixture 1: (1) tert-butylbenzene, (2) iso-butylbenzene and (3) n-butylbenzene; mixture 2: (4) 1,3,5-trichlorobenzene, (5) 1,2,4-trichlorobenzene and (6) 1,2,3-trichlorobenzene; mixture 3: (7) o-nitrotoluene, (8) m-nitrotoluene and (9) p-nitrotoluene; mixture 4: (10) m-chloronitrobenzene, (11) p-chloronitrobenzene and (12) o-chloronitrobenzene.

3.4 Preliminary applications in real samples

On account of the high resolving capability of CB7–OV column for isomers, we made preliminary applications for detection of minor isomer impurities in commercial reagent samples of analytical grade by using 1 mL of one reagent sample and performing the GC separations under the indicated conditions. Fig. 6 shows the chromatograms for separations of the reagent samples of o-nitrotoluene, m-nitrotoluene, o-butylbenzene and tert-butylbenzene, respectively. Table 6 lists the approximative assay results for the main component and minor isomer impurity for each of the samples measured by peak area normalization method. As shown, the measured purity for each sample was in good accordance with its labelled purity with isomer impurity in the range of 0.08–1.48%. These preliminary results demonstrate the feasibility of CB7–OV column for the separation and detection of minor isomers in real samples.

Fig. 6 Preliminary applications of CB7–OV capillary column for the detection of isomer impurities in four real reagent samples of (a) o-nitrotoluene, (b) m-nitrotoluene, (c) iso-butylbenzene and (d) tert-butylbenzene, respectively. Temperature programs: 50 °C for 1 min to 160 °C at 15 °C min⁻¹ for (a) and (b); 40 °C for 1 min to 160 °C at 10 °C min⁻¹ for (c) and (d).

Table 6 Results for the detection of minor isomer impurities in the commercial reagent samples on the CB7–OV column

Sample	Labelled purity	Measured purity	Isomer impurity	Content
o-Nitrotoluene	>99%	99.86%	m-Nitrotoluene	0.14%
m-Nitrotoluene	>97%	98.02%	o-Nitrotoluene	1.48%
iso-Butylbenzene	>99%	99.58%	n-Butylbenzene	0.22%
tert-Butylbenzene	>98%	99.82%	n-Butylbenzene	0.08%

---

4 Conclusions

The proposed strategy in this work achieved the highest column efficiency ever reported for CBn columns and also greatly enhanced the resolving ability of the binary stationary phase due to the possible synergistic effect of CB7 with OV-210. CB7–OV column exhibits advantageous separation performance for structural and positional isomers and preferential retention for alkanes, aromatics and H-bonding donors. Moreover, it has good column repeatability and thermal stability. The results in this work demonstrate the potential of CB7–OV column in GC separations of analytes with high resemblance.

Acknowledgements

The authors are grateful for the financial support by the National Natural Science Foundation of China (21075010, 21575013) and the 111 Project B07012 in China.

References

1. J. Lagona, P. Mukhopadhyay, S. Chakrabarti and L. Isaacs, Angew. Chem., Int. Ed., 2005, 44, 4844–4870.
2. D. Bardelang, K. A. Udachin, D. M. Leek, J. C. Margeson, G. Chan, C. I. Ratcliffe and J. A. Ripmeestert, Cryst. Growth Des., 2011, 11, 5598–5614.
3. S. J. Barrow, S. Kasera, M. J. Rowland, J. Barrio and O. A. Scherman, Chem. Rev., 2015, 115, 12320–12406.
4. A. C. Bhasikuttan, H. Pal and J. Mohanty, Chem. Commun., 2011, 47, 9959–9971.
5. X. L. Ni, X. Xiao, H. Cong, Q. J. Zhu, S. F. Xue and Z. Tao, Acc. Chem. Res., 2014, 47, 1386–1395.
6. S. D. Choudhury, N. Barooah, V. K. Aswal, H. Pal, A. C. Bhasikuttan and J. Mohanty, Soft Matter, 2014, 10, 3485–3493.
7. X. L. Ni, X. Xiao, H. Cong, L. L. Liang, K. Cheng, X. J. Cheng, N. N. Ji, Q. J. Zhu, S. F. Xue and Z. Tao, Chem. Soc. Rev., 2013, 42, 9480–9508.
8. B. C. Pemberton, R. Raghunathan, S. Volla and J. Sivaguru, Chem.–Eur. J., 2012, 18, 12178–12190.
9. P. Zhang, S. J. Qin, M. L. Qi and R. N. Fu, J. Chromatogr. A, 2014, 1334, 139–148.
10. T. Sun, N. N. Ji, M. L. Qi, Z. Tao and R. N. Fu, J. Chromatogr. A, 2014, 1343, 167–173.
11. L. Wang, X. G. Wang, M. L. Qi and R. N. Fu, J. Chromatogr. A, 2014, 1334, 112–117.
12. X. G. Wang, M. L. Qi and R. N. Fu, J. Chromatogr. A, 2014, 1371, 237–243.
13. Y. Z. Wang, M. L. Qi and R. N. Fu, RSC Adv., 2015, 5, 76007–76013.
14. L. Y. Ma, S. M. Liu, L. Yao and L. Xu, J. Chromatogr. A, 2015, 1376, 64–73.
15. M. L. Qi and D. W. Armstrong, Anal. Bioanal. Chem., 2007, 388, 889–899.
16. J. O. Grisales, P. J. Lebed, S. Keunchkarian, F. R. González and C. B. Castells, J. Chromatogr. A, 2009, 1216, 6844–6851.
17. P. J. Lebed, S. Keunchkarian, J. Osorio Grisales and C. B. Castells, J. Chromatogr. A, 2014, 1324, 198–206.
18. A. Curat, S. Tisse, A. Andrieu, N. Bar, D. Villemin and P. Cardinael, Chromatographia, 2014, 77, 1671–1681.
19. J. H. Zhang, S. M. Xie, L. Chen, B. J. Wang, P. G. He and L. M. Yuan, Anal. Chem., 2015, 87, 7817–7824.
20. J. H. Zhang, S. M. Xie, B. J. Wang, P. G. He and L. M. Yuan, J. Chromatogr. A, 2015, 1426, 174–182.
21. E. F. Barry and R. L. Grob, Columns for Gas Chromatography: Performance and Selection, John Wiley & Sons, New Jersey, 2007.

---