Deactivation by polysiloxane and phenyl containing disilazane : a 29Si CP-MAS NMR study after the formation of polysiloxane chains at the surface

(1)

Deactivation by polysiloxane and phenyl containing disilazane

: a 29Si CP-MAS NMR study after the formation of

polysiloxane chains at the surface

Citation for published version (APA):

Hetem, M. J. J., Rutten, G. A. F. M., Ven, van de, L. J. M., Haan, de, J. W., & Cramers, C. A. M. G. (1988). Deactivation by polysiloxane and phenyl containing disilazane : a 29Si CP-MAS NMR study after the formation of polysiloxane chains at the surface. HRC & CC, Journal of High Resolution Chromatography and

Chromatography Communications, 11(7), 510-516. https://doi.org/10.1002/jhrc.1240110703

DOI:

10.1002/jhrc.1240110703 Document status and date: Published: 01/01/1988 Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

providing details and we will investigate your claim.

(2)

Deactivation by Polysiloxane and Phenyl Containing

Disilazane: A29Si CP-MAS NMR Study after the Formation

of

Polysiloxane Chains at the Surface

M. Hetem*, G. Rutten, L. van de Ven, J. de Haan, and C. Cramers

Eindhoven University of Technology, Dept. of Chemistry, Lab. Instrumental Analysis, P. 0. Box 513, 5600 MB Eindhoven. The Netherlands

Key

Words:

Capillary GC Deactivation 29Si-NMR

Summary

A high degree of deactivation of glass and fused-silica capillary column walls is attainable by means of high temperature silylation (HTS) with or without a preceding leaching process. HTS

with a phenyl containing disilazane, diphenyltetramethyldisila- zane (DPTMDS), and polydimethylsiloxane (PDMS) are studied onCab-O-Sil,afumedsilica,asamodelsubstrate.Using29SiCP-

MAS NMR, it was shown that no dirnethylsiloxane chains were formed upon silylation with DPTMDS under different conditions of humidity and stoichiometry at 377OC. With DPTMDS deactivation it is possible that amino trisiloxy silane groups areformed, these groups add extra activity to the surface. Silylation with a PDMS, OV 101, at various temperatures between 30Oo42O0C did show that dimethylsiloxane chains were bonded at the surface. Using the 29Si CP-MAS NMR technique with variable contact times to reveal siloxy group mobility, the degradation of dimethylsiloxane chains at the surface was studied. PDMS degradation at an optimal temperature gives a more effective diminuation of the silane activity caused by chemical reaction with thesilanolgroupsandtheeffectivescreeningof theremain- ing silanol groups with anchored polydimethylsiloxane chains and small cyclodimethylsiloxane ring structures at the surface.

1 Introduction

A number of silylation reagents have been reported in the

literature for the deactivation of the innerwall of fused silica columns in gas chromatography under widely different conditions at temperatures above 3OO0C, resulting in deactivating layers of various nature [1,2]. Here, two different widely applied deactivation mechanisms are studied on Cab-0-Sil, a vitreous quartz. The two silylation methods compared are a deactivation by a phenyl containing disilazane, 1,3-diphenyl-l, 1,3,3-tetramethyIdisilazane [DPTMDS], introduced by Grob et a/. [3] and a deactivation by degradation of polydimethylsiloxane [PDMS], introduced by Schomburg et al. [4,5].

The mechanism of the HTS reaction with phenyldisilazanes was studied previously in our laboratory [6-81. The main result was that at relatively mild, dry conditions (T= 35OOC)

benzene was released selectively from the silane moieties at the surface.

In 1981, LiY F. eta/, [9]studiedtheuseofshort(cyclic)alkyl polysiloxanes, which not only reacted with active silanol groups at the surface, but also masked the remaining sila- no1 groups through the formation of short, bridged, surface- anchored dimethylsiloxane chains. The formation

of

loop structures by the dimethylsiloxanes at the silica surface might therefore improve the deactivation.

Recently, Reiher

[lo]

suggested that dimethylsiloxane chains were formed at the silica surface from phenyl containing silazanes by reaction of initially formed aryldimethylsiloxy silanes with water liberated from the surface (see Figure 1). The dimethylhydroxysiloxysilane would then react further with the excess of silazane forming short chains consisting of a few dimethylsiloxane units and terminated with an aryldimethylsiloxy silane. In this way, the mechanisms of HTS with phenyl-containing disilazanes and polysiloxanes would be intertwined. It was assumed that the presence of water at the surface favors the formation of these (short) dimethylsiloxane chains, which would explain both the relatively easy loss of benzene and the high critical surface tension using an excess of DPTMDS. In order to test this hypothesis, the silylation of Cab-0-Sil with DPTMDS was carried out at 377OCfor 16 h underdiffer- ent conditions of humidity and stoichiometry: two different

Figure 1

Aryldimethylsiloxy siloxanes react with water liberated from thesurface according to [lo].

(3)

Deactivation by Polysiloxane and Phenyl Containing Disilazane

Cab-0-Sit batches were used, an extremely dry Cab-0-Sit and a humid Cab-0-Sil conditioned over asaturated potas- sium bromide solution for a period of several months. The latter batch of Cab-0-Sil contained 5.5 w/w% of water. The amounts of DPTMDS added to both the Cab-0-Sil batches were either stoichiometric amounts of silylation reagents or ten times the stoichiometric ratio.

For reasons of comparison also silylation by polysiloxane degradation of PDMS at a number of temperatures, between 300°C and 42OOC for 16 h, on dried Cab-0-Sil is studied. With 29Si CPMAS NMR and 29Si Bloch pulse MAS NMR techniques the anchoring of (short) dimethylsiloxane chains to the surface of the fumed silica was studied with the explicit aim to investigate whetheror not larger, moreor less mobile chains are present at the surface of the silylated Cab-0-Sil.

2 Experimental

2.1 Materials

The Cab-0-Sil M5 (Cabot Corp., Tuscola, 111, USA) was a gift from Heybroek & Co's Handelsmij. N.V., Amsterdam, NL. The specific surface area of grade M5 is, according to the manufacturer, 200

k

25 m2/g. A value of 200 m2/g was used in this work. 1,3-Diphenyl-l,1,3,3-tetramethyIdisilazane was obtained from Ffuka A.G., 9470 Buchs, Switzerland (purum) and the polydimethyl siloxane, OV-101, was obtained from Ohio Valley Specialty Chem. Co., Ohio, USA. Other chemicals and solvents used were all analytical grade from E. Merck, Darmstadt, FRG.

2.2 Pretreatment of the Cab-0-Sil

The Cab-0-Sil M5 was ignited at 720°C, and rehydrated as described before [6]. Part of the Cab-0-Sil wasdriedfurther over P2O5 in a vacuum desiccator for several weeks. Part of

Table 1

Cab-0-Sil samples. Experimental conditions, weight % carbon.

the Cab-0-Sit was conditioned in airwith 84% relative humi- dity over a saturated solution of KBr. This Cab-0-31 batch contained 5.5% (w/w) water. This equals 15.3 pmol/m2.

2.3 Silylation of Cab-0-Sil

2.3.1 DPTMDS

About 0.4 g of Cab-0-Sil was placed in a quartz glass reaction ampoule (length 20 cm, i.d. 1 cm, wall thickness 1 mm). A constriction was drawn in the middle of the tube and the required amount of reagent was added with a syringe. The tube was evacuated twice (because the volatility of the deactivation reagens, the ampoule was meanwhile cooled in dry ice) and filled with nitrogen to atmospheric pressure. After the final evacuation the ampoule was sealed to a volume of about 8 ml.

The amount of silylation reagent was calculated according to the densest attainable surface concentration of tri- methylsiloxy groups formed on hydrated silica by Stober [ l l ] : 4.7 pmol/m2. Therefore, 0.47 mmol DPTMDS for the stoichiometric reaction and 4.7 mmol DPTMDS for the excess reaction were added per gram dry Cab-0-Sil. For the humid Cab-0-Sil batch these amounts were calculated to be 2.0 mmol DPTMDS for the stoichiometric reaction and 6.23 mmol DPTMDS for the excess reaction, assuming that 2 mol H20 reacts with 1 mol DPTMDS.

2.3.2 PDMS

Cab-0-Sil was coated with 1.96 g OV-101 per gram, which corresponds to a layer of 10 nm. The OV-1 01 was dissolved in pentane and the corresponding amount of Cab-0-Sil was added. The pentane was evaporated under reduced pressure in a rotary evaporator. The resulting coated Cab-0-Sil was dried in a vacuum desiccator over P2O5 for several weeks. About 0.4 g of the coated Cab-0-Sil sample was placed in quartz glass reaction ampoules and vacuum sealed as described for DPTMDS.

Reaction Weight Yo

Sample Pretreatment Reagents Amountlg

No. Cab-0-Sil Ca b-0-Si I temp. ("'2.) carbona)

1 dried over P205 DPTMDS 0.134 377 3.04

3 above K Brb) DPTMDS 0.57 377 4.30

4 above K Br DPTMDS 1.78 377 4.30

6 dried over P2O5 ov-101 1.96 300 10.26

8 dried over P2O5 ov-1 01 1.96 380 6.50

9 dried over P205 ov-1 01 1.96 420 3.01

a) Weight % carbon determined after silylation and pentane washing.

b) Samples conditioned over K Br contained 5.5 wlw % of water.

c) The PDMS used is OV-101,

&

= 30000 g. mol-1.

2 dried over P205 DPTMDS 1.342 377 5.33

5 dried over P2O5 ov-101") 1.96 110

7 dried over P205 ov-101 1.96 340 9.33

(4)

After sealing the ampoules were wrapped in aluminum foil, placed in a well ventilated oven and heated to the required temperatures for 16 h. After reaction the ampoules were opened and the contents washed twice with toluene and twice with pentane. The Cab-0-Sil samples were then dried overnight in a vacuum oven at 1 10°C. Table 1 lists the sily- lated Cab-0-Sil samples.

2.4 Elemental

Carbon

Analysis

The carbon content of the silylated Cab-0-Sil was obtained with a Perkin ElmerAnalyzermodel240(Perkin ElmerCorp., Avondale, CT, USA). Tungsten oxide was added to the deactivated silica as a catalyst.

2.5 NMR Measurements

The 29Si and 13Csolid state NMR spectra were obtained on a Bruker CXP 300 spectrometer at 59.63 MHz and at 75.48 MHz respectively. The samples were spun at ca. 3.5 kHz using aluminum oxide rotors of the standard Bruker double bearing type.

In cross-polarization experiments variable contacts were used with contact times of 1,2,4, and 8 ms for 29Si and 1 ms for 13C spectra. Acquisition times of 10 ms (29Si) and 29 ms (1%) and pulse interval times of 1 s and 2 s respectively, were applied. Typically, 12,000 fid's (29Si) and 3,000 fid's (13C) were accumulated in 1Kdata points, zero-filled to 8K prior to Fourier transformation. The spectral width was 20 kHz. Line broadening used is 20 Hz prior to zero filling and Fourier transformation.

3 Results and Discussion

3.1 Earlier Work

Earlier, models for silica deactivation by means of polysiloxane degradation and by reaction with, e.g., D4 (octa- methylcyclotetrasiloxane), HMDS, TPSA, and DPTMDS

have been studied on Cab-0-Sil and with GC in our laboratory [6-81. All chemical shifts were assigned before. The chemical shifts most relevant to this paper are collected in

Table 2. The corresponding cyclosiloxanes at the surface

are presented in Figure 2.

Unfortunately the CPMAS NMR technique does not distin- guish between cyclosiloxane structures with four or five units. From polysiloxane degradation experiments at temperatures between 4OOOC and 50O0C, as published before [7], it can be concluded, largely on 29Si NMR chemical shift arguments and on cross polarization behav- ior, that rather small fragments of polydimethyldisiloxy- silanes are formed at the surface. However, no terminal groups were detected and only a relatively small, narrow signal in the dimethyldisiloxysilane region near -22 ppm was noticed, assigned on the basis of chemical shift arguments to longer, mobile polydimethylsiloxanes.

3.2 DPTMDS Silylation

Samples 1-4 (Table 1) were prepared with DPTMDS at 377OC and 29Si CP-MAS NMR spectra were recorded, con- tact time 4 ms. From the 29Si NMR spectra (Figure 3) it can be concluded that a variety of silylation products is formed

0; c y c l a p e n t a n l o x a n e

Me Me

0; Cyclotetrall loxane

Table 2

Siloxane/silane functionality, notation and typical 29Si chemical shift.

Figure 2

Small ring structures with dimethyldisiloxysilanes at the surface of the silica.

Chemical Topological Code (spectral) (network) functionality functionality Typical shift (ppm downfield from TMS.) phenyldimethylsiloxysilane dimethyl hydroxysiloxysilane dimet hyldisiloxysilane paired dimethyldisiloxysilane poly(dimet hyldisiloxysilane) aminotrisiloxysilane di hydroxysiloxane hydroxysiloxane tetrasiloxysiloxane 1 1 2a) 2b) 2 3 2 3 4 + 2 - 4 - 8

-

16

-

22 - 88 - 91

-

101

-

110 a) dimethyldisiloxysilane has a bidentate linkage to the surface forming a cyclotrisiloxane

(5)

1

I \

0 - 5 0 -100 ppm.

Figure 3

%i CPMAS NMRspectra of DPTMDSsilylatedCab-0-Sil. Samples 1 to4. CT = 4 ms, LB = 20 Hz and N, = 12000.

at the silica surface with chemical shifts between +2 and

-18

ppm from TMS. For sample 4 also a discernable signal at

-88

ppm, assigned to aminotrisiloxysilane

[8],

is detected. This illustrates a large difference in product distribution influenced by reaction conditions similar to those used for DPTMDS deactivation of capillary columns. The primary coupling product of DPTMDS, phenyldimethylsiloxysilane (MIP,

+2

ppm) is found for all four samples, although the relative amounts vary. The other silylation products in the region between

-4

and

-18

ppm are mainly assigned to bifunctional dimethyldisiloxysilanes as listed in Table 2.

Wthastoichiometricamount of DPTMDS(samp1es

1

and3) relatively large amounts of D2’ cyclotetrasiloxane and possibly larger siloxane ring systems are formed. When an excess of DPTMDS is added (samples 2 and 4) instead of these Dzr cyclosiloxane structures, substantial amounts of D2 cyclotrisiloxanes are formed, especially in the presence of water. An explanation of this phenomenon could be that the silylation of the surface starts with covering the silica with the primary coupling product phenyldimethylsiloxysi- lane. The primary phenyldimethylsiloxysilanes at the silica surface will than presumably react further with the remaining neighbouring silanol groups with formation of

dimethyldisiloxysilanes (D2 cyclotrisiloxane) and evolution of benzene. This reaction is promoted by the presence of water at the surface (sample 4).

An alternative reaction occurs between two neighboring

phenyldimethylsiloxysilanes resulting in a shared siloxane

bridge (D2’ cyclotetrasiloxane). The latter reaction predominates, when a stoichiometric ratio of DPTMDS is used for deactivation at 377OC.

The loss of the phenyl group is confirmed by 13C CPMAS NMR (Figure 4) and elemental carbon analysis (Table 1). Presumably, steric hindrance caused

by

a high concentra- tion of primary attached phenyldimethylsiloxysilanes partly prevents formation of larger cyclosiloxane systems. The presence of aminotrisiloxysilane ( T d ) at the surface of the Cab-0-Sil sample 4 was discussed extensively by Van

de Ven et a/.

[8].

Aminotrisiloxysilane is formed in the

presence of water especially when an excess of DPTMDS is added to the silica. The changes of the critical surface tension reported

[lo]

could very well be caused by the

aromatic C

1 -

H I SI- C - H

7

Figure 4

WCPMAS NMRspectraof DPTMDSsilylatedCab-0-Sil.Samples 1 to4. CT = 1 ms, LB = 20 Hz and N, = 3280.

(6)

93

A

M.P D..

I 1

Figure 5

Varying contact times, 29Si CPMAS NMR spectra of DPTMDS silylated Cab-0-Sil, sample 4, LB = 20 Hz and N, = 12000.

aminotrisiloxysilane groups formed at the fused silica surface. This amino group, however, adds an extra activity to the surface and decreases the deactivation.

Variation of the contact times, for sample 4 (Figure 5) shows that none of the signals near -1 6 ppm is enhanced upon increasing contact time, only thesignal at lowest field, assigned to phenyldimethylsiloxysilanes shows this effect, in line with the known tendency of trialkylsilyl groups (alkyl

5 ethyl). The growth of the quaternary signal at -1 10 ppm here is due to indirect crosspolarization via protons at least four bonds away [12], not to larger mobilities. The sharp signal at -22 ppm, characteristic of dimethydisiloxysilane chains anchored at the Cab-0-Sil surface is missing. It can be concluded that no mobile dimethyldisiloxysilanes are formed at the surface of the Cab-0-Sil. Dimethylhydroxysi- loxysilanes (D,) are probably present at the surface of all four samples, but no longer chains are formed even in the presence of water and large amounts of reagents added. This contradicts the reaction scheme proposed by Reiher (see Figure 1).

3.3 PDMS Silylation

Samples 6-9 (Table 1) were prepared with OV 101 at tem- peratures between 3OOOC and 420OC. 29Si CP-MAS NMR spectra of these samples together with Cab-0-Sil coated

/ \

9 4

Figure 6

2aSi CPMAS NMR spectra of PDMS silylated Cab-0-Sil at elevated temperatures, sample 5 to 9, CT = 4 ms, LB = 20 Hz and N. = 2000.

with OV 101 and dried at 11 0°C in vacuum (sample 5), were recorded with a contact time Of 4 ms (Figure 6). In the 29Si

NMR spectra one sharp signal at -22 ppm occurs. This signal, as mentioned before, is assigned to polydimethylsiloxane (DZ”, Table 2). The signal at -1 8 to -20 ppm, which increases with increasing reaction temperatures up to 38OoC, corresponds to the formation of degradation products of polydimethylsiloxane chains, existing of dimethyldisiloxysilanes surface paired (or D2’ cyclotetrasiloxane) and larger, anchored dimethylsiloxane ring structures. At temperatures between 300 and 340°C, practically no chain degradation is found. Decreasingly less polydi- methylsiioxanes are found after silylation between 380 and 420°C. However, there is no indication that D2 cyclotrisiloxanes or dimethylhydroxysiloxysilane end groups are formed. The loss of larger dimethylsiloxane chains is confirmed by elemental carbon analysis (Table 1).

A difference in the maximum for D2’ cyclosiloxane structures is recorded between PDMS and DPTMDS silylation. The maximum for the CPMAS NMR signal representing D2’

(7)

CT.= 2 ms.

CT.= 4 ms.

C T . = a m s .

cyclosiloxane after PDMS deactivation is situated 3 ppm upfield from D2' cyclosiloxane products after DPTMDS deactivation. This shift in maximum could be explained by deshielding (change of *%i NMR signal position towards lower field) of dimethyldisiloxysilanes in larger ring structures (up to five or six units) or can be caused by the first units of the polydimethylsiloxane chains close to the surface. These anchored units do not possess the high mobility of a liquid as polymer chain dimethylsiloxane units, while their NMR shielding is still influenced by the chemical bonding at the surface.

Variation of contact times clearly indicates that thesignal at -22 ppm belongs to rather mobile silane groups (in the

lo4

Hz

range), in contrast to other signals in the same region (Figure 7 ) . This is in complete agreement with the earlier assignments of this signal (see above). The average mobility of the D2" moieties at the surface of sample 9 is reduced compared to sample 6. This is shown by the signal at -22 ppm, which is less influenced by spectrometric contact time variation. The spectrum of sample 5 (Figure 6) indicates that practically no surface bonding has taken place at 1 10°C, under vacuum. The corresponding 29Si NMR signal could hardly be obtained by crosspolarization but was relatively easy to excite by conventional pulse methods (Figure

a),

as usual in measuring NMR spectra for very mobile groups. By pulse excitation it is shown that at

42OOC less than 5 percent of the original polydimethylsiloxanes are left at the surface.

At increasing temperatures the signals of hydroxysiloxane (Q3) and dihydroxysiloxane

(a,)

are decreasing rapidly (Figure 5) and so is the silanol activity of the surface. The silanol groups left will be covered with anchored polydimethylsiloxane chains, and small cyclosiloxane loops.

Figure 7

Varying contact times, 2QSi CPMAS NMRspec- tra of PDMS silylated Cab-0-Sil, samples 6 and 8, LB = 20 Hz and N, = 2000. D; 94 Q4 n

-

---a

0; I1 -50 -1011 ppm. Figure 8

28Si Bloch pulse MAS NMR spectra of samples 6 and 9, N, = 5500.

From this it can be concluded that PDMS degradation at an optimal temperature gives an effective diminution of the silanol activity and adds no extra activity to the surface.

Acknowledgment

The investigations were supported (in part) by The Netherlands Foundation for Chemical Research (SON) with financial aid from the Netherlands Organization for the Advancement of Pure Research (ZWO). Our grateful acknowledgments is due to Mrs. Denise Tjallema for her expeditious handling of the manuscript.

(8)

References [7] G. Rutten, J. de Haan, 1. van de Ven, A. van de Ven, H. van

6. J. Xu, Ph. D. Thesis, Free University, Amsterdam, March 1987.

M. 1. Lee and 6. W. Wright, J. Chromatogr. 184 (1980) 235- 31 5.

K. Grob, G. Grob, and K. Grob Jr., HRC & CC 2 (1 979) 31 -35. G. Schomburg, H. Husmann, and H. Borwitzky, Chromato- graphia 12 (1979) 651-6133.

G. Schomburg, H. Husmann, S. Ruthen, and M. Herrair, Chro- matographia 15 (1982) 599-610.

G. Rutten, A. van de Ven, J. de Haan, L. van de Ven, and J. Rijks, HRC & CC 7 (1984) 607-614.

Cruchten, and J. Rijks, HRC & CC 8 (1985) 664-672. [8] 1. van de Ven, G. Rutten, J. A. Rijks, and J. W. de Haan, HRC & [9] Li Yu-Fu, Xia Yong-Xia, Xu Dong-Peng, and Li Guang-Liang, J.

Polym. Sci., Polym. Chem. Ed. 19 (1981) 3069-3077.

[lo]

T. Reiher, HRC&CC 10 (1987) 158-159. [l I] W. Stober, Kolloid Zeitschr. 149 (1956) 39-46.

1121 G. E. Maciel and D. W. Sindorf, J. Am. Chem. SOC. 102 (1 980)

Ms received: January 4,1988 Accepted by PS: January 21,1988 CC 9 (1 986) 741 -746.