Translate this page into:
Hollow silica microspheres as the stationary phase for thin layer chromatographic separation of a model mixture
⁎Corresponding author. Address: School of Materials Science and Engineering, University of Jinan, No. 106, Jiwei Road, Jinan 250022, Shandong, PR China. Tel./fax: +86 531 82769106. liusq_ujn@hotmail.com (Shiquan Liu)
-
Received: ,
Accepted: ,
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.
Peer review under responsibility of King Saud University.
Abstract
Hollow silica microspheres (HSMs) with different structures have been synthesized using sacrificed hard template route combined with multiple sol–gel silica shell coating steps. The synthesized HSMs were characterized by SEM and N2 sorption measurements and employed as the stationary phase in thin layer chromatography. Thin layers of HSMs were coated on glass slides and used to separate a model mixture of methyl red and dimethyl yellow. The conditions to achieve the best chromatographic separation efficiency of the synthesized HSMs were optimized. The results show that the organic mixture could be well-separated using the mixture of cyclohexane/toluene/ethanol as the mobile phase. Distilled water was the best mixing agent for the preparation of the thin layer plates. Activation of the plates at 105 °C for 1 h improved the separation efficiency. Under the optimized conditions, the effect of the microstructures of HSMs on the separation efficiency was analyzed. It is shown that the separation efficiency mainly depends on the thickness of the silica shells and the pore size gradient of the nanopores inside the shells. A comparison test shows that HSMs as a stationary phase are advantageous over commercial silica gel in an easy preparation of homogeneous TLC thin plates.
Keywords
Thin layer chromatography
Hollow silica microspheres
Organic mixture
Nanopores
Shell thickness
1 Introduction
Thin layer chromatography (TLC) is used to separate organic mixtures, especially as a simple method to primarily determine how many different organic substances exist in unknown mixtures. TLC has now been developed into a trace analysis technology because the method is very simple, fast and low cost. And only a small volume of sample is needed. In the literature, TLC has been reported in the analysis of drugs (Ali et al., 2012), biological samples (Tan, 2011) and foods (Yang et al., 2007). For example, Mostafa (2010) used the TLC method for a quantitative determination of paracetamol both in its bulk powder and in pharmaceutical dosage forms in the presence of degradation product. Sharma and Sharma (2016)) studied the simultaneous estimation of Irinotecan HCl in their dosage forms by the spectrophotometric method combined with the TLC densitometric method. Baghdady et al. (2013) separated ezetimibe and atorvastatin calcium on silica gel plates using the TLC-densitometry method.
TLC relies on many different factors, including the stationary phase (sorbent), the mobile phase (developing agent), and the interaction between the substances to be separated and both phases. TLC is performed on glass sheets, plastic or aluminum foils on which the sorbent materials are spread. Therefore, it is also important to have TLC plates with suitable thickness and mechanical strength (Brenner and Niederwieser, 1967; Pant, 2009; Santiago and Strobel, 2013). The separation of different components in organic mixtures via TLC mainly depends on differences in the solubility of the components in the mobile phase and the adsorption capacity of the components in the stationary phase. Each component has a different desorption rate in the mobile phase and a different adsorption rate in the stationary phase. Thus, the migration rate and distance of each component following the mobile phase are different. Substances strongly adsorbed to the sorbent will move slowly and those weakly interacted with the sorbent will move fast (Fried and Sherma, 1999). As a result, each component will appear at different positions on the thin layer plate, forming different spots or ribbons. Then, different components in the mixture can be differentiated (Wang et al., 2003).
It is very important to select a suitable stationary phase when using TLC to separate mixtures. Silica gel Tantawy et al. (2012), Chavhan et al. (2017), Weng et al. (2003), Richter et al. (2008), and Parent et al. (2006) has widely been adopted as the stationary phase. Other reported stationary phases include 200–300 mesh waste glass powder (Pant, 2009) and cellulose (Tao et al., 2011). To evaluate the performance of a substance as a new stationary phase in TLC, real and model mixtures can be used as compounds to be separated. A model mixture consists of already known components that would be suitable for a primary evaluation of a new stationary phase. However, differences in the structure and properties of the components to be separated should be considered. If a stationary phase can separate two or more components with small differences from a mixture, the stationary phase could be a good candidate in real applications. The mixture of methyl red and dimethyl yellow could be a good model mixture, in which both components only have a difference in the carboxylic group. This small difference in the molecular structure of these two components is helpful for judging the TLC separation performance of a stationary phase. In addition, the distinct colors of these two components facilitate an easy visualization in the TLC operation (Santiago and Strobel, 2013).
In the present work, hollow silica microspheres (HSMs) with different microstructures have been synthesized and used as the stationary phase in TLC to separate the model mixture of methyl red and dimethyl yellow. In addition, the influence of the structure of HSMs on the TLC separation performance is analyzed. The results indicate that HSMs with finely tuned microstructures could be a good candidate for TLC.
2 Experimental section
2.1 Raw materials
Micrometer-sized polystyrene (PS) microspheres (diameter of 3–4 μm) were synthesized using the procedures presented in Anthony, 1990. Tetraethoxysilane (TEOS), ethanol (EtOH) and toluene were from Sinopharm Chemical Reagent Co., Ltd, China; dodecyltrimethylammonium bromide (DTAB), tetradecyltrimethylammonium bromide (TTAB) and cetyltrimethylammonium bromide (CTAB) were from Shanghai Pure Reagent Co., Ltd; ammonia solution (NH3·H2O), methyl red, dimethyl yellow, and cyclohexane were from Laiyang Kangde Chemicals Co., Ltd, Tianjin Damao Chemical Reagent Factory, Tianjin Kemiou Chemical Reagent Co., Ltd, Tianjin Fuyu Fine Chemicals Co., Ltd, respectively. All the reagents were of analytical grade and used as received.
2.2 Preparation and characterization of HSMs with different structures
HSMs were prepared via the dual-template sacrificed route. Details can be found in our previous publications (Wei et al., 2011). Briefly, silica shells were coated on the PS cores with the assistance of surfactants. After the removal of both the PS and surfactant templates via the calcination of the core–shell samples at 550 °C for 6 h, hollow silica microspheres were obtained. To tune the microstructure of the microspheres, one, two or three layers of silica were coated respectively in different runs. Thus, silica shells with different thicknesses were obtained. The second and the third silica coating were performed on the composite PS-silica microspheres obtained in the previous run(s). In the meantime, surfactants with different alkyl chain-lengths were applied in different coating runs so that pores with different sizes could be generated in different silica layers after the removal of the assembled surfactants. The samples were encoded according to the layers of silica coatings and the surfactant used in each layer. Thus, sample D corresponds to one layer of silica prepared with the addition of DTAB. Samples DD, DT, and DC consist of two layers of silica, in which the first layer was formed with the addition of DTAB while the second layer with DTAB, TTAB, and CTAB, respectively. Sample DTC is composed of three layers of silica, in which DTAB, TTAB and CTAB, were used respectively in the first, second and third layer of silica coating. The core template and shell forming media are listed in Table 1.
Core template/g
Media used for the silica shell coating
PS
DTAB–SiO2@PS
TTAB–SiO2/DTAB–SiO2@PS
EtOH/ml
H2O/ml
CTAB/g
TTAB/g
DTAB/g
NH3·H2O/ml
TEOS/ml
D
4
19
11
0.605
4.3
1.2
DT
4
19
11
0.8
4.3
0.6
DTC
4
19
11
0.8
4.3
0.6
The morphology and shell thickness of the samples were characterized on a Scanning Electron Microscope (SEM) (JEOL JSM-5510). The samples were sputtered with a thin film of gold. N2 sorption isotherms were recorded at 77 K on a Quantachrome Autosorb-iQ-C automated gas adsorption system. The samples were degassed at 80 °C for 0.5 h, then at 200 °C for 10 h. The Brunauer–Emmett–Teller (BET) method was used to calculate surface areas. The pore volumes were taken at the relative pressure p/p0 close to 1. For a simple comparison, all the pore size distribution (PSD) curves were obtained using the Barret–Joyner–Hallender (BJH) model based on the adsorption data.
2.3 TLC separation experiments
Normal glass slides were carefully cleaned and dried. Then, 0.025 g hollow silica microspheres were dispersed in a mixing agent through ultrasonic dispersion. The mixing agent used in the study included absolute ethanol, or distilled water or starch solution (25 g/l). The dispersion was dropped and distributed on the glass plates. After being dried in the air over night, the plates were then heated at 105 °C for 1 h. After the solvent was evaporated, a plate was put into a bottle containing the mobile phase. Then, the mobile phase carrying the components in the model mixture moved upward. To optimize the separation conditions, the plates were quickly taken out when the front of the mobile phase moved 4.5 cm from the starting point. To compare the migration rates of the components in the plates, the plates were quickly removed after they were put in the mixture for 15 min. For comparison, 0.16 g commercial TLC silica gel (Qingdao Haiyang Chemical Co., Ltd, details of the silica gel can be found in the following section) was also used to prepare plates, whose separation performance was also tested under the above optimized conditions.
3 Results and discussion
3.1 Characterization of the synthesized HSMs
Fig. 1 shows the SEM images of the synthesized products. They all have a spherical morphology. The broken particles reveal that the products are hollow. The microspheres are around 3–4 μm in diameter, close to the size of the PS beads which templated the hollows of the microspheres. However, the wall thickness differs from one sample to another. Based on the cross-sections of the broken microspheres (see the insets in Fig. 1), the thicknesses of samples D, DT, and DTC are 65, 95, and 140 nm respectively.
SEM images of HSM samples (a) D, (b) DT, and (c) DTC.
The N2 isotherms and PSD curves of samples D, DT and DTC are shown in Fig. 2. Sample D shows a type I isotherm, indicating it is microporous. Sample DTC shows a type IV isotherm, characteristic of a mesoporous structure. However, the isotherm of sample DT is in-between Type I and II isotherms, suggesting the existence of both micropores and mesopores (Rouquerol et al., 1999). The calculated surface areas, pore volumes and pore diameters of the HSMs are listed in Table 2. It can be seen that the surface area, pore volume and pore diameter of the HSMs increase with increasing shell thickness. The data has shown the overall textural characteristics of the HSMs, however, do not reflect the detailed microstructures of the silica shells. In fact, these increases in the data are due to the increasing alkyl chain lengths of the surfactant used in the first, second and the third layer of the silica coatings. The size of the nanopores is mainly determined by the size of the surfactant micelles. Theoretically, the nanopores gradually decrease in size from the third outermost layer to the first inner layer depending on the chain length of the template used in the formation of the relevant layers. This feature of the pore size is called “pore size gradient” in the following text. However, due to the small differences in the pore size, both the sorption measurement and the transmission electron microscopy which are frequently used to characterize the pore size cannot differentiate this pore size gradient.
Nitrogen sorption isotherms (A) and pore size distribution curves (B) of the HMS samples (a) D, (b) DT, and (c) DTC.
Sample
Surface area (m2/g)
Pore volume (ml/g)
Pore diameter (nm)
D
643
0.585
1.69
DT
996
0.657
1.88
DTC
1766
0.797
2.19
TLC silica gel
274
0.798
9.61
3.2 Influence of the types of the mobile phases
One of the key factors influencing the TLC separation is the type of the mobile phase, whose polarity, volatility as well as the solubility of the components to be separated in the mobile phase have to be considered (Wang et al., 2003). Taking sample DT as an example of the stationary phase, three different mixtures consisting of (a) cyclohexane/toluene (3:4 (v:v)); (b) cyclohexane/toluene/ethanol (1.5:5:0.5 (v:v:v)); (c) toluene/ethanol (5:2 (v:v) were tested as the mobile phase. The separation results are compared in Fig. 3. It can be seen from Fig. 3a that the organic mixture can be separated when using cyclohexane/toluene (3:4) as the mobile phase. However, methyl red hardly moved with the mobile phase, whereas the migration distance of dimethyl yellow is large. The result indicates that the desorption of methyl red from the mobile phase was quicker and the interaction of methyl red with the stationary phase was stronger than those of dimethyl yellow (Fried and Sherma, 1999). When cyclohexane/toluene/ethanol (1.5:5:0.5) was the mobile phase, components in the mixture were well-separated, forming two obvious ribbons with a distance of 7 mm on the plate (Fig. 3b). In addition, the migration distance of dimethyl yellow is larger than that shown in Fig. 3a, suggesting that the mobile phase of cyclohexane/toluene/ethanol has a weaker adsorption on the stationary phase than cyclohexane/toluene (3:4). In the case of using the mixture of toluene/ethanol (5:2) as the mobile phase, both methyl red and dimethyl yellow moved simultaneously with the mobile phase. Thus, the separation is very poor. This is due to the strong polarity of the mobile phase (Santiago and Strobel, 2013), which has a strong absorption of the organics.
Separation of the mixture of methyl red and dimethyl yellow using different mixtures as mobile phases (a) cyclohexane/toluene (3:4 (v:v)), (b) cyclohexane/toluene/ethanol (1.5:5:0.5 (v:v:v)), and (c) toluene/ethanol (5:2 (v:v)).
3.3 Influence of the activation treatment
The as-prepared thin layer plates which were dried at room temperature are called non-activated plates. The dried plates could be activated by heating treatment in an oven at 105 °C for 1 h. Using cyclohexane/toluene/ethanol (1.5:5:0.5 (v:v:v) as the mobile phase and sample DT as the stationary phase, the influence of the activation treatment on the separation performance was investigated. From Fig. 4, it can be seen that both the non-activated and the activated plates can separate the mixture of methyl red and dimethyl yellow. However, the activated plate shows slightly improved performance, indicated by the smaller separated color points and a larger distance between the ribbons. It is known that there are silanol groups on the surface of the nanopores inside the HSMs (Vansant et al., 1995). Water can be easily adsorbed on the HSMs, decreasing the activity of the microspheres (Brenner and Niederwieser, 1967). On the other hand, the polarity of the mobile phase increases due to the presence of water.
Separation of the mixture of methyl red and dimethyl yellow using (a) non-activated plate and (b) activated plate.
3.4 Influence of the mixing agents
Ethanol, or distilled water or starch solution (25 g/l) was used as the mixing agent in the preparation of the thin layer plates. The mixture of cyclohexane/toluene/ethanol (1.5:5:0.5) and sample DTC were used as the mobile and the stationary phases, respectively. The plates were activated at 105 °C for 1 h. The separation results are compared in Fig. 5. It can be seen that water is the best mixing agent. The plate prepared with the addition of distilled water separated the mixture into clear colorful ribbons at two different positions on the plate. In the case of the plate prepared with starch solution, the separated ribbons show lag tails. Ethanol is the least favorable mixing agent. This may be due to the weak interaction between the ethanol molecules, which prevented the formation of compact HSM layers on the glass slides.
Separation of the mixture of methyl red and dimethyl yellow using plates prepared with different mixing agents (a) EtOH, (b) H2O, and (c) starch solution.
3.5 Evaluation of HSMs with different structures as the stationary phase
Under the optimized conditions, thin layer plates of HSMs with different structures were used to separate the model mixture. The separation pictures are shown in Fig. 6. The migration distances of the components in the mixture were measured and compared in Table 3. The two separated colorful ribbons displayed on the plates indicate that our synthesized hollow microspheres as the stationary phase can separate the mixture very well. The data in Table 3 demonstrate that both the migration rate and the distance between two separation ribbons increase from sample D to DT and DTC. Thus, the separation efficiency is in the order of DTC > DT > D. We also proved that uncalcined samples, which were not porous, could not separate the mixture at all. Either component in the mixture did not move along with the mobile phase in the plates prepared with the non-porous composite microspheres. Therefore, the nanopores and hollows play a significant role in the separation of the organic mixture (Santiago and Strobel, 2013).
Separation of the mixture of methyl red and dimethyl yellow using plates prepared with HSM samples as the stationary phase (a) D; (b) DT; (c) DTC, and (d) DS under the same separation conditions.
Stationary phase
D
DT
DTC
CTD
DC
CD
TLC silica gel
Migration distance of/mm
Methyl red
15
17
23
20
15
17
9
Dimethyl yellow
21
27
35
31
23
24
37
Difference in the migration distance/mm
6
10
12
11
8
7
28
Taking the shell thickness and the pore parameters (surface area, pore volume, and pore diameter) of the HSMs into consideration, the increase in the separation efficiency in the order of DTC > DT > D could be due to either the increasing shell thickness or the increasing pore parameters or both of them. To further clarify this, another HSM sample encoded with DS was synthesized and tested in the separation of the mixture. First, this sample was synthesized following the same procedure as sample D. Then the obtained DTAB–SiO2@PS composite microspheres were used as the core. The second layer of silica coating was performed in a medium with reactant ratios of EtOH:H2O:NH3·H2O:TEOS = 4.17:1:0.47:0.08. The shell thickness of the sample is 110 nm, larger than that of sample D (65 nm). As one can see, no surfactant was used in the coating of the second layer of silica. Therefore, the pores in the first (inside) layer of silica were templated by DTAB and the pores in the second (outside) layer of silica were generated during the gel-drying process. The pore distribution inside the silica shells is gradient, from small to large inwards. The overall surface area (227 m2/g), pore volume (0.183 ml/g) and pore size (0.57 nm) of sample DS are significantly smaller than those of sample D. Fig. 6d shows that sample DS could also separate the mixture and the migration rates and distances of the separated organics are larger than those in the case of sample D. The results suggest that the pore parameters are not the factor determining the separation performance of the hollow silica microspheres. Instead, the increasing shell thickness and the pore size gradient, which influence the retention of the components in the absorbent (Fried and Sherma, 1999), contributed to the observed enhanced performance of sample DS. This also explains why the HSMs with three layers of silica have better performance than those with two layers of silica (see data presented in Table 3 for samples DC, CD and CTD).
Finally, commercial TLC silica gel was used in the same TLC under the optimized conditions. The N2-sorption isotherm, PSD curve and the optical image of the silica gel particles are shown in Fig. 7(a–c), respectively. The surface area, the pore volume and the BJH pore size are 274 m2/g, 0.798 ml/g and 9.6 nm. These data are also listed in Table 2. Compared with the HSM sample DTC, the silica gel has a much smaller surface area and larger pore size. In addition, the gel particles are not uniform in size, ranged from a few to about 15 μm (Fig. 7c). The separation figure (Fig. 7d) and the migration distance data (see Table 3) show that the commercial silica gel as the stationary phase has a much larger difference in the migration distance of the organics, suggesting a better separation performance than our HSM samples. The fast migration rate of dimethyl yellow might be attributed to the significantly large pore size and small surface area of the silica gel. However, our stationary phase is advantageous over the commercial silica gel in several aspects. First, the amount of HSM samples needed for preparing a plate is much less (only one-sixth) than that of the commercial silica gel used. This is mainly due to the much smaller packing density of our HSMs compared with the commercial silica gel. In addition, the silica gel is irregular in shape and not uniform in size, the plate made of the silica gel has to be prepared much thicker. Otherwise, the silica gel layer on the plate is inhomogeneous and very easy to peel off from the glass slides (Pant, 2009).
The N2-sorption isotherm (a), PSD curve (b), optical image (c) and separation image (d) of a commercial silica gel.
4 Conclusions
Hollow silica microspheres (HSMs) with different microstructures were synthesized via the dual-template route with multiple sol–gel silica coatings. Thin layer chromatography (TLC) plates were prepared with the synthesized HSMs as stationary phases. The conditions for achieving a good separation performance of the synthesized sorbent were optimized. Under the optimized conditions, TLC plates prepared with hollow silica microspheres with a large shell thickness and pore size gradient in the silica shells as the stationary phase show the best performance in separating the model mixture consisting of methyl red and dimethyl yellow. The HSMs as the stationary phase for TLC are advantageous over commercial silica gel in preparing TLC plate with homogeneous and thin chromatographic layers with a much less amount.
Acknowledgments
We gratefully acknowledge the financial support from the National Science Foundation of China (50972052).
References
- Arab. J. Chem.. 2012;5:235-240.
- J. Polym. Sci. Part A. 1990;28:2485-2500.
- J. Adv. Res.. 2013;4:51-59.
- Methods Enzyme. 1967;11:39-59.
- Arab. J. Chem.. 2017;10:S825-S830.
- Thin Layer Chromatography (fourth ed.). New York: Marcel Dekker; 1999.
- J. Saudi Chem. Soc.. 2010;14:341-344.
- Waste Manage.. 2009;29:2040-2041.
- Appl. Surf. Sci.. 2006;252:6746-6749.
- J. Chromatogr. B. 2008;867:233-237.
- Adsorption by Powders and Porous Solids. London: Academic Press; 1999.
- Methods Enzyme. 2013;533:303-324.
- Arab. J. Chem.. 2016;9:S1368-S1372.
- Food Fermen. Technol.. 2011;47(2):64-68.
- Adv. Res. 2012 http://dx.doi.org/10.1016/j.jare.2012.05.004
- Food Control. 2011;22:1154-1157.
- Characterization and Chemical Modification of the Silica Surface. Amsterdam: Elsevier; 1995.
- Basic Experimental Chemistry. Beijing: Science Press; 2003. (in Chinese)
- Micro. Meso. Mater.. 2011;138:17-21.
- J. Pharm. Biomed. Anal.. 2003;33:85-93.
- Phys. Testing Chem. Anal. Part B Chem. Anal.. 2007;43(3):229-232.