Sie sind auf Seite 1von 7

G Model

CHROMA-358499;

No. of Pages 7

ARTICLE IN PRESS

Journal of Chromatography A, xxx (2017) xxx–xxx

IN PRE SS Journal of Chromatography A, xxx (2017) xxx–xxx Contents lists available at ScienceDirect Journal

Contents lists available at ScienceDirect

Journal of Chromatography A

journal homepage: www.elsevier.com/locate/chroma

A journal homepage: www.elsevier.com/locate/chroma Behavior of macroporous vinyl silica and silica monolithic

Behavior of macroporous vinyl silica and silica monolithic columns in high pressure gas chromatography

Adrien Maniquet b , Nicolas Bruyer a , Guy Raffin a , Franck Baco-Antionali b , Claire Demesmay a , Vincent Dugas a , Jérôme Randon a,

a Univ Lyon, CNRS, Université Claude Bernard Lyon 1, Ens de Lyon, Institut des Sciences Analytiques, UMR 5280, 5 rue de la Doua, F-69100 Villeurbanne, France b IFP Énergies nouvelles – établissement de Lyon, Expérimentation Procédés, Rond-point de l’échangeur de Solaize, BP3, 69360 Solaize, France

a r

t i

c l e

i n f o

Article history:

Received 12 April 2017 Received in revised form 27 April 2017 Accepted 2 May 2017 Available online xxx

Keywords:

Hybrid silica monoliths Silica monoliths Gas chromatography High pressure

a b s t r a c t

80% vinyltrimethoxysilane-based hybrid silica monoliths (80-VTMS), which have been initially devel- oped for separation in reversed-phase liquid chromatography, have been investigated in high pressure gas chromatography separations (carrier gas pressure up to 60 bar) and compared to silica monolithic columns. The behavior of both silica and 80-VTMS monolithic columns was investigated using helium, nitrogen and carbon dioxide as carrier gas. The efficiency of 80-VTMS monolithic columns was shown to vary differently than silica monolithic columns according to the temperature and the carrier gas used. Carrier gas nature was a significant parameter on the retention for both silica and vinyl columns in relation to its adsorption onto the stationary phase in such high pressure conditions. The comparison of retention and selectivity between 80-VTMS monoliths and silica was performed under helium using the logarithm of the retention factor according to the number of carbon atoms combined to Kovats indexes. The very good performances of these columns were demonstrated, allowing the separation of 8 compounds in less than 1 min.

© 2017 Elsevier B.V. All rights reserved.

1. Introduction

Monolithic capillary columns result of an in-situ synthesis of

porous material inside microchannels, and organic monoliths were introduced in separation science in the 70’s as an alternative to packed column [1]. Such organic monoliths were characterized by

a relatively simple experimental design for each of the large set

of monomers used leading to a wide variety of surface chemistry.

If these columns based on an organic backbone can also be used

in GC for even samples containing a large amount of water, their major drawbacks were related to their low thermal stability and their low permeability [2–5]. On the other side, inorganic monoliths can also be produced based on silica (SiO 2 ) [6] or transition metal oxides such as zirconium dioxide (ZrO 2 ) [7,8], or titanium diox- ide (TiO 2 ) [9,10] for separation purpose. Silica monolithic columns have been introduced by Tanaka group in the 90’s for liquid chro- matography [11], and in comparison with organic monoliths, they

are mechanically and thermally more resistant [12]. Even if the sur- face properties can be tuned by several ways depending on the

Corresponding author. E-mail address: randon@univ-lyon1.fr (J. Randon).

http://dx.doi.org/10.1016/j.chroma.2017.05.002

0021-9673/© 2017 Elsevier B.V. All rights reserved.

surface chemistry required for a specific LC application [13], the investigation of inorganic monolith in gas separation are really few, mainly because it requires specific instrumentation to work with high carrier gas pressure [3,14]. Light hydrocarbon mixtures from C1 to C6 are the main targeted applications for inorganic monolithic columns in relation to the adsorption mechanism involved in GC for these compounds, but the inorganic surface chemistry available up to now was limited to silica leading to only one specific selectivity. A new type of hybrid monolithic columns (80-VTMS: 80% vinyltrimethoxysilane-based hybrid silica monoliths) have been developed and manufactured in our lab and have shown multi- ple benefits over their grafts counterparts when used in reverse phase liquid chromatography [15,16]. These columns were made of vinyltrimethoxysilane and showed better hydrophobic properties, and higher chemical stability than silica columns with conven- tional methyl groups. Moreover, among a wide range of hybrid materials synthetized, vinyl groups are good candidates to be used later as a generic platform for further specific modifications [15,16]. The behavior of these columns has been studied in high pressure gas chromatography and compared to conventional silica columns already used in GC [17]. The influence on the carrier gas nature, pressure and column temperature has been studied on retention,

G Model

CHROMA-358499;

No. of Pages 7

ARTICLE IN PRESS

2

A. Maniquet et al. / J. Chromatogr. A xxx (2017) xxx–xxx

Table 1 Summary of silica and 80-VTMS monolithic columns (geometry and performances).

Column number

Internal diameter (SEM)/ m

Pore diameter (SEM)/ m

Length/cm

Permeability B 0 (GC)/m 2

(t RC2 /t RC1 )

75 C,

H min of C 2 , 75 C, CO 2 / m

20 bar, CO 2

SiO 2 (M1-1) SiO 2 (M3-1) SiO 2 (M3-2)

68

1.9

16.4

3.1 × 10 14 6.3 × 10 14 6.4 × 10 14 5.7 × 10 13 7.4 × 10 13 3.2 × 10 13

1.25

15

70

1.9

18.6

1.22

14

61

1.9

22.8

1.23

18

80-VTMS

(L1-1)

65

2.0

35.1

1.33

81

80-VTMS

(L2-1)

67

2.0

59.5

1.34

99

80-VTMS

(L3-1)

67

2.0

26

1.32

86

efficiency and selectivity of light hydrocarbons gases on the 80- VTMS columns.

2. Materials and methods

2.1. Chemical and reagents

Fused-silica capillaries (75 m I.D., 375 m O.D.) were obtained from Cluzeau (France). Tetramethoxysilane (TMOS), vinyltrimethylsiloxane (VTMS), nitric acid, formamide, methanol, ethanol, propanol, isopropanol, isooctane and decane were pur- chased from Sigma–Aldrich (France), toluene and dichloromethane from Fisher Chemical (France), pentane and undecane from Acros Organics (France) and octane were from Prolabo (France). All aqueous solutions were prepared using 18 M water (Millipore, Molsheim, France). Nitrogen and carbon dioxide were from Air Liquide (France). Helium and sample mixtures were from Messer (France): mix A [methane (998 ppm), ethane (970 ppm), ethylene (994 ppm), acetylene (1019 ppm), propane (987 ppm), cyclo- propane (1016 ppm), n-butane (989 ppm), 1-butene (989 ppm), carbon monoxide (996 ppm), carbon dioxide (1008 ppm), n - pentane (100 ppm), n-hexane (100 ppm), hydrogen (98.6 ppm) in nitrogen) and mix B [hydrogen (15 mol%), carbon monox- ide (15 mol%), carbon dioxide (15 mol%), methane (15 mol%) and ethane (15 mol%) in nitrogen]. Propene was obtained from IFPEN Lyon (France).

2.2. Preparation of 80-VTMS silica capillary monolithic columns

Silica monolithic columns have been prepared by sol-gel process from tetramethoxysilane [17] while 80-VTMS columns have been prepared from a mixture of TMOS/VTMS (20:80, w/w) according to the previously published protocol [13,14], briefly reported here. A total of 3.2 g of this mixture was added to 1.24 g of nitric acid (1 M) and 2.8 g of formamide and stirred at 0 C for 5 min. Then, the mix- ture was kept at 40 C for 48 h for gelation and ageing. The mixture was then flushed into fused-silica capillaries by applying a con- trolled pressure using a Micro Fluidic Control System (MFCS) from Fluigent (France). The monoliths were washed for 12 h with water under hydrodynamic flow and dried at 100 C for 12 h. A scanning electron microscope (SEM) Neoscope JCM-5000 from JEOL (France) was used to measure the internal diameter of the silica capillaries and to estimate the pore size diameter of the monoliths.

2.3. 80-VTMS and silica capillary monolithic columns

Several columns were made from different batches and their characteristics are reported in Table 1. Columns name are rep- resented as follow: SiO 2 (M [batch number-column number]) or 80-VTMS (L [batch number-column number]). Columns were compared in terms of internal diameter, pore diameter, length, permeability, relative retention time of C 2 over C 1 (at 75 C, car- rier gas CO 2 20 bar, from mixtures A and B) and H min (minimum plate height). Results were very similar when comparing differ- ent parts of the same column and inter batch columns. This set

of homogenous columns from different batches has been used in order to be sure that the reported results were representative of silica and 80-VTMS monolithic column behaviors. With an appar- ent pore size around 2 m, the permeability was expected to be not far from 10 13 m 2 which was the permeability range observed. For capillary columns produced with the same material, permeability values are rather similar (standard deviation 30%). Furthermore, permeability in porous media is also related to geometrical charac- teristics such as connectivity, tortosity and pore circularity which is dependant on the synthesis protocol. As a consequence, 80-VTMS columns were ten times more permeable, less efficient than silica in the best conditions, and the difference in surface chemistries leads to a shift in relative retention time of ethane compared to methane.

2.4. Gas chromatography

An Agilent Technologies 7820 gas chromatograph (GC) with a split/splitless injector and a flame ionization detector (FID) was lent by SRA Instruments (France), and gas injection valve and high pressure regulators have been added to the GC. Valves, tubing, fer- rules and nuts were in stainless steel 316L from Swagelok (France). Pressure regulators were bought from Messer (France) and Air Liquide (France). Pressures gauges were obtained from Keller AG (Switzerland). The injection valve for gas samples was fromVICI Valco (Switzerland): the model used has an internal sample loop of 0.06 L and a pressure limit of 1000 psi (69 bar). Valve actuation was performed with helium. Liquid samples were injected at 120 C using an on-line liquid injector which could be use at elevated pres- sure. Split injection was designed using open capillary tubing with controlled flow resistance based on internal diameter and length of the capillary. The system validation and hydrodynamic char- acterization of the chromatographic device have been reported in a previous paper. Analysis of light hydrocarbons mixtures in gaseous samples were conducted with an oven temperature int the range 25 C 120 C. Retention factors have been determined using methane as hold up time marker above 75 C (below 75 C methane appeared slightly retained) [17].

3. Results and discussion

3.1. Influence of the temperature and the nature of the carrier gas

on 80-VTMS column efficiency in comparison with silica monoliths

Van Deemter’s curves of ethane from mixture B have been mea- sured for 80-VTMS and silica monolithic columns for each carrier gas (He, N 2 and CO 2 ) for inlet pressure ranging from 10 to 50 bar and for temperature ranging from 25 C to 120 C. Fig. 1 shows the Van Deemter’s curves for helium (left) and car- bon dioxide (right) on 80-VTMS column. These curves highlight the temperature effect on efficiency of 80-VTMS columns and its dependence on the carrier gas. The temperature effect on column efficiency appeared to be much more important with helium than with carbon dioxide. In this case, effect of temperature on column efficiency seems to be mainly related to a decrease of the mass transfer resistance parameter on the stationary phase when tem-

G Model

CHROMA-358499;

No. of Pages 7

ARTICLE IN PRESS

A. Maniquet et al. / J. Chromatogr. A xxx (2017) xxx–xxx

600

500

400

300

200

100

0

H / μm 25°C 50°C 75°C 100°C 80-VTMS under helium 0 5 10 15 20
H / μm
25°C
50°C
75°C
100°C
80-VTMS under helium
0
5
10
15
20
25
30

u / cm.s -1

400

350

300

250

200

150

100

50

0

H / μm 25°C 50°C 75°C 100°C 80-VTMS under CO 2 0 5 10 15
H / μm
25°C
50°C
75°C
100°C
80-VTMS under CO 2
0
5
10
15
20

u / cm.s -1

3

Fig. 1. Van Deemter plots of ethane from mixtureB (sample pressure at1.3 bar) for temperatures from 25 C to 120 C under helium (left) and carbon dioxide (right) on 80-VTMS (L2-1)monolithic column.

3

2

2

1

1

0

H 50 /H 120

A : Temperature influence 2.1 SiO 2 1.8 80-VTMS 1.6 1.2 1.2 1.1
A : Temperature influence
2.1
SiO 2
1.8
80-VTMS
1.6
1.2
1.2
1.1

M1-1

M3-1

M3-1

L1-1

L2-1

Column number

L3-1

5

4

4

3

3

2

2

1

1

0

H He /H CO2

B : Carrier gas influence 4.1 3.6 SiO 2 3.4 80-VTMS 2.0 1.6 1.5
B : Carrier gas influence
4.1
3.6
SiO 2
3.4
80-VTMS
2.0
1.6
1.5

M1-1

M3-1

M3-2

L1-1

L2-1

Column number

L3-1

Fig. 2. H 50 C /H 120 C under helium (left) andH He /H CO2 at 75 C (right) for silica and 80-VTMS monolithic columns.

Table 2 H min and optimal velocity for 80-VTMS (L2-1) monolithic column under different carrier gases (He, N 2 and CO 2 ) for different temperatures: 25, 50, 75, 100 and 120 C

Carrier gas

25

C

50 C

75 C

100 C

120 C

/Temperature

 

H min / m

u opt /cm s 1

H min / m

u opt /cm s 1

H min / m

u opt /cm s 1

H min / m

u opt /cm s 1

H min / m

u opt /cm s 1

He

316

6

220

8

153

10

129

11

121

13

N 2

226

6

152

6

117

6

96

8

90

8

CO 2

174

7

110

7

99

8

87

8

82

8

perature increases. However, whatever the carrier gas (He, N 2 or CO 2 ), the minimum height equivalent to a theoretical plate was always obtained at high temperature, with the lower value (slightly lower than 100 m) obtained with CO 2 as carrier gas (Table 1). The effect of temperature and carrier gas nature on minimum height equivalent to a theoretical plate has also to be compared to the one observed on silica monoliths, and was reported on Fig. 2. The gain in efficiency according to a temperature shift from 50 C to 120 C was first compared under helium (Fig. 2-A). Tempera- ture has a significant effect on the minimum height equivalent to a theoretical plate with helium for 80-VTMS columns (gain factor 1,8 ± 0,3), whereas temperature has no major effect on the effi- ciency of silica columns (gain factor 1,1 ± 0,1). So, high temperature

has to be used for 80-VTMS columns in order to reach the maximum efficiency of the column. The effect of the carrier gas nature on effi- ciency was also studied for silica and 80-VTMS columns (Fig. 2-B) at an intermediate temperature of 75 C and so the ratio of H min values obtained with He/CO 2 was compared for both silica and 80-VTMS columns (higher is this ratio, better is carbon dioxide compared to helium). The effect of the nature of the carrier gas led to a substantial gain of a factor 3,7 ± 0,4 on the efficiency of silica monolithic columns and it was less pronounced with 80- VTMS columns with only a factor 1,7 ± 0,3. This difference could be explained from a much higher micro/meso-porosity on silica compared to 80-VTMS columns leading to a more significant pore obstruction of micro-pore by carbon dioxide on silica. This could

G Model

CHROMA-358499;

No. of Pages 7

ARTICLE IN PRESS

4

A. Maniquet et al. / J. Chromatogr. A xxx (2017) xxx–xxx

1

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

k N2 or CO2 /k He

He N 2 - 120°C CO 2 - 120°C N 2 - 75°C CO 2
He
N 2 - 120°C
CO 2 - 120°C
N 2 - 75°C
CO 2 - 75°C
0
5
10
15
20
25
30
35

ΔP / bar

Fig. 3. k N2 /k He and k CO2 /k He for ethane from mixture B (sample pressure 1 bar) versus P at 75 C and 120 C on column 80-VTMS (L3-1).

lead to a lower mass transfer resistance and as a consequence to

a lower minimum height equivalent to a theoretical plate. Nitro-

gen acts as an intermediary carrier gas between helium and carbon dioxide on efficiency as reported in Table 2 with H min and u opt for

80-VTMS monolithic column for different carrier gases and several temperatures.

Silica based columns are very efficient (more than 10 000 plates per meter) compared to the value reported in literature and could be appropriate stationary phases for the analysis of highly volatile compounds. Concerning 80-VTMS columns, they reach their max- imum efficiency at elevated temperature and so would be more appropriate for analysis of heavier compounds. For both silica and 80-VTMS columns, best efficiencies are reach when carbon dioxide

is used as carrier gas, with a significant influence on silica columns

as highlighted in Fig. 2 leading to more than 60 000 plates per meter.

3.2. Influence of the temperature and nature of the carrier gas on

80-VTMS columns retention

The influence for various carrier gas on the retention and selec- tivity was previously reported on several monolithic columns [2,3,14,17–26] in high pressure gas chromatography but appeared

to be stationary phase dependent. Such behavior was here inves- tigated on 80-VTMS columns for C 2 at different temperatures and pressures using helium, nitrogen and carbon dioxide as carrier gas (see ref 15 for silica columns). When helium was used as a carrier gas, helium pressure had no influence on retention factor what- ever the temperature, so these retention factors can be used as

a reference for other carrier gases (nitrogen and carbon dioxide). Relative retention factors were plotted on Fig. 3 where retention factors using nitrogen and carbon dioxide were normalized over the retention factor observed with helium at the same temper- atures. Whatever the oven temperature, retention factor of light hydrocarbons decreased when nitrogen or carbon dioxide pressure increased. The molecules of the carrier gas being adsorbed on the stationary phase surface reduced the number of available adsorp- tion sites accessible for analyte molecules leading to a decrease of the apparent retention factor. Such reduction of the number of available sites for analyte molecules is less important when the temperature increases or when the carrier gas pressure decreases

Table 3 Kovats retention indexes for several compounds under helium on SiO 2 (M3-1) and 80-VTMS (L3-1) columns at 120 C.

 

SiO 2 (M3-1)

80-VTMS (L3-1)

Ethylene

210

200

Propene

325

300

1-butene

436

400

Acetylene

300

200

Cyclopropane

348

339

Dichloromethane

638

555

Toluene

852

800

Methanol

941

555

Ethanol

1026

651

due to a lower adsorption of carrier gas molecules on the monolithic surface.

3.3. Comparison of the retention and selectivity of monolithic

columns in silica and 80-VTMS

In order to study the influence of vinyl groups on the retention and selectivity of 80-VTMS columns, compared to silica columns, the logarithm of the retention factor of several liquid and gaseous compounds was plotted versus the number of carbon atoms for SiO 2 and 80-VTMS columns (Fig. 4). These analysis have been performed under helium to avoid the effect of the carrier gas pressure on reten- tion and at 120 C allowing methane to be monitored as the holdup time marker [3]. Kovats retention indexes have been reported in Table 3 to compare in an easier way the change in retention and selectivity between the two stationary phases. Retention of linear alkanes increased from silica to 80-VTMS columns, and such phenomenon could be related to the vinyl groups which enhance the surface hydrophobicity or to a larger surface area of the stationary phase. So both silica and 80-VTMS columns have been synthesised as macro monoliths (6 mm diam- eter, 2 cm height) and their specific surface areas were measured. The specific surface area of silica columns (550 m 2 g 1 ) was higher than for 80-VTMS columns (159 m 2 g 1 ). Assuming the same spe- cific surface area ratio in capillary columns, the higher alkane retention on 80-VTMS columns, despite à lower surface of adsorp- tion, appeared to be related to the more hydrophobic vinyl groups.

G Model

CHROMA-358499;

No. of Pages 7

ARTICLE IN PRESS

3.0

2.5

2.0

1.5

1.0

0.5

0.0

-0.5

-1.0

3.0

2.5

2.0

1.5

1.0

0.5

0.0

-0.5

-1.0

A. Maniquet et al. / J. Chromatogr. A xxx (2017) xxx–xxx SiO 2 log(k) Alkanes
A. Maniquet et al. / J. Chromatogr. A xxx (2017) xxx–xxx
SiO 2
log(k)
Alkanes
Alkenes
Acetylene
Alcohols
Aroma c rings
Dichloromethane
C 11 H 24
C 2 H 5 OH
C 10 H 22
C 9 H 20
CH 3 OH
Cyclo-C 7 H 8
C 8 H 18
C 7 H 16
CH 2 Cl 2
C 6 H 14
C 5 H 12
C 4 H 8
Cyclo-C 3 H 6
0
1
2
3
4
5
6
7
8
9
10
11
12
C 3 H 6
C 4 H 10
C 2 H 2
C 3 H 8
C 2 H 4
C 2 H 6
Carbon number log(k) 80-VTMS C 11 H 24 C 10 H 22 C 9 H
Carbon number
log(k)
80-VTMS
C 11 H 24
C 10 H 22
C 9 H 20
Cyclo-C 7 H 8
C 8 H 18
C 3 H 7 OH
C 7 H 16
C 2 H 5 OH
C 6 H 14
CH 3 OH
C 5 H 12
CH 2 Cl 2
C 4 H 8
Cyclo-C 3 H 6
C 4 H 10
0
1
2
3
4
5
6
7
8
9
10
11
12
C 3 H 8
C 3 H 6
C 2 H 2
C 2 H 6
C 2 H 4

Carbon number

5

Fig. 4. Log(k) versus the number of carbon atoms under helium on SiO 2 (M3-1) and 80-VTMS (L3-1) columns at 120 C.

Concomitantly, alkene retention increased but with a lower extend compared to alkanes so their Kovats indexes decreased, even leading to a co-elution between ethylene and ethane, propene and propane, and 1-butene and butane at 120 C. Acetylene Kovats index decreased more significantly ( I K 100) compared to alkenes ( I K 20-50). Acetylene which was clearly separated to other C2 compounds on silica, now appeared closer to ethane and ethene. This phenomenon was similar to all groups of alkanes and alkenes with the same number of carbon atoms which formed clusters on 80-VTMS columns, as display on chromatograms per- formed at 25 C (Fig. 5) with a shift in peak positions also observed for butane and butene. Kovats indexes of alcohols decreased dramatically on 80-VTMS columns. And more interestingly, alcohols which showed a strong peak asymmetry on silica columns due to the presence of silanol groups on the surface, appeared with more symmetric shape on 80-VTMS columns probably due to a reduced number of residual

silanol groups on the surface. For another polar compound such as dichloromethane, retention also decreased as Kovats indexes varied from 638 to 555.

4. Conclusion

Characterization of several silica and 80-VTMS monolithic capillary columns has confirmed the good repeatability and repro- ducibility for intra-batch columns which reach efficiency at the optimum velocity in the order to 60 000 plates per meter for silica. Differences of efficiency between the two types of columns were dependent of temperature and nature of carrier gas. The influence of the nature of the carrier gas on efficiency revealed a gain by a fac- tor 4 on silica columns and a factor 2 and 80-VTMS columns when CO 2 was used instead of He. The adsorption of CO 2 on the station- ary phase plays also an important role on the retention whatever the column type used, leading to a strong reduction of retention

G Model

CHROMA-358499;

No. of Pages 7

ARTICLE IN PRESS

6

A. Maniquet et al. / J. Chromatogr. A xxx (2017) xxx–xxx

FID response /pA C2 SiO 2 4.0 Ethane Méthane 3.5 C3 Ethylène 3.0 Propane 2.5
FID response /pA
C2
SiO 2
4.0
Ethane
Méthane
3.5
C3
Ethylène
3.0
Propane
2.5
C4
Acétylène
Cyclopropane
2.0
n-butane
1.5
1.0
1-butène
0.5
0.0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Time /min
14
Méthane
Ethane /
C2
12
80-VTMS
Ethylène
10
Acétylène
8
C3
6
Propane
Cyclopropane
C4
4
n-butane
2
1-butène
0
0 0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1

Time /min

Fig. 5. Chromatogram of mixture A at 20 bar and 25 C using CO 2 as carrier gas on silica and 80-VTMS columns.

when carrier gas pressure increased. Temperature effects, from 50 to 120 C, are negligible on silica columns and allow a substantial gain (factor 2) on 80-VTMS columns. The difference of selectivity between the two stationary phases has shown the formation of clusters according to their number of carbon atoms on 80-VTMS columns. These columns are complementary and allow fast analysis of a wide range of volatile and rather apolar compounds.

Acknowledgments

The authors are grateful to SRA Instruments for providing the conventional GC and installing the injection valve. They also thank Tiffany Vittori and Mikaïl Aksoy for conducting preliminary tests during their internships.

References

[5] D. Sykora,´

monoliths: a novel format for capillary gas chromatography stationary phases, Macromol. Mater. Eng. 275 (2000) 42–47. [6] K. Nakanishi, S. Naohiro, Phase separation in gelling silica–organic polymer solution: systems containing poly(sodium styrenesulfonate), J. Am. Ceram. Soc. 74 (1991) 2518–2530. [7] D.C. Hoth, J.G. Rivera, L.A. Colón, Metal oxide monolithic columns, J. Chromatogr. A 1079 (2005) 392–396. [8] J. Randon, S. Huguet, A. Piram, G. Puy, C. Demesmay, J.-L. Rocca, Synthesis of zirconia monoliths for chromatographic separations, J. Chromatogr. A 1109 (2006) 19–25.

85 (2004) 5595–5597.

1275–1280.

[12] G. Puy, C. Demesmay, J.L. Rocca, J. Iapichella, A. Galarneau, D. Brunel, Electrochromatographic behavior of silica monolithic capillaries of different

skeleton sizes synthesized with a simplified and shortened sol-gel procedure, Electrophoresis 27 (2006) 3971–3980.

[13] O. Núnez,˜

231–252.

[14] A. Kurganov, Monolithic column in gas chromatography, Anal. Chim. Acta 775 (2013) 25–40. [15] R. El-Debs, F. Cadoux, L. Bois, A. Bonhommé, J. Randon, V. Dugas, C. Demesmay, Synthesis and surface reactivity of vinylized macroporous silica monoliths: one-pot hybrid versus postsynthesis grafting strategies, Langmuir

31 (2015) 11649–11658.

1016/j.chroma.2016.07.011.

capillary columns, Russ. J. Phys. Chem. 80 (2006) 1135–1140. [19] T.N. Gvosdovich, A.V. Kiselev, Y.I. Yashin, Study of the effect of the nature of the carrier gas on retention volumes and differential molar changes of internal energy during adsorption on some porous polymers, Chromatographia 11 (1978) 137–140. [20] G.S.M. Rabbani, M. Rusek, J. Janák, Gas chromatographic investigation of some ethylvinylbenzene polymers using different carrier gases, J. Gas Chromatogr. 6 (1968) 1966–1967. [21] A. Karmen, I. McCaffrey, R.L. Bowman, Use of carbon dioxide as carrier gas in gas chromatography, Nature 193 (1962) 575–576. [22] V.G. Berezkin, I.V. Malyukova, V. Alishoev, The influence of the carrier gas on retention in capillary gas-solid chromatography, J. High Resolut. Chromatogr.

G Model

CHROMA-358499;

A. Maniquet et al. / J. Chromatogr. A xxx (2017) xxxxxx

7

[23] A.V. Kozin, A.A. Korolev, V.E. Shiryaeva, T.P. Popova, A.A. Kurganov, The influence of the natures of the carrier gas and the stationary phase on the separating properties of monolithic capillary columns in gas adsorption chromatography, Russ. J. Phys. Chem. 82 (2008) 276–281. [24] A.A. Korolev, V.E. Shiryaeva, T.P. Popova, A.A. Kurganov, Enthalpy-entropy compensation effect on adsorption of light hydrocarbons on monolithic stationary phases, J. Sep. Sci. 34 (2011) 2362–2369.

[25] A.A. Korolev, E.V. Shiryaeva, T.P. Popova, A.A. Kurganov, Thermodynamic parameters of the sorption of light hydrocarbons on monolithic capillary columns in gas chromatography, Russ. J. Phys. Chem. A 81 (2007) 1136–1140. [26] A.A. Korolev, V.E. Shiryaeva, T.P. Popova, A.A. Kurganov, Polar monolithic capillary columns: analysis of light hydrocarbons, Russ. J. Phys. Chem. A 87 (2013) 120–124.