Anal. Chem. 2005, 77, 7862-7871
Molecular Dynamics Simulations of Alkylsilane Stationary-Phase Order and Disorder. 2. Effects of Temperature and Chain Length Katrice A. Lippa,*.† Lane C. Sander,† and Raymond D. Mountain‡
Analytical Chemistry Division and Physical and Chemical Properties Division, Chemical Sciences and Technology Laboratory, National Institute of Standards and Technology, 100 Bureau Drive, Gaithersburg, Maryland 20899
In an effort to elucidate the molecular-level structural features that control shape-selective separations, we have investigated the molecular dynamics of chromatographic models that represent both monomeric and polymeric stationary phases with alkylsilane length and temperature conditions analogous to actual materials of low to high shape selectivity. The structural characterization of these models is consistent with previous experimental observations of alkyl chain order and disorder: alkyl chain order increases both with alkyl chain length and with reduced temperature. Models that represent shape-selective reversed-phase liquid chromatography (RPLC) phases possess a significant region of distal end chain order with primarily trans dihedral angle conformations; the extension of these ordered regions into the phase increases with an increase in chain length. Models with extended chain length (C30) possess a higher degree of conformational order and are relatively insensitive to changes in surface coverage, bonding chemistry, and temperature. Chromatography models of various chain lengths and over a temperature range that represents highly shape-selective RPLC stationary phases all contain a series of well-defined and rigid cavities; the size and depth of these “slots” increase for the C30 models, which may promote the enhanced separations of larger size shape-constrained solutes, such as carotenoids. Alkyl-modified microparticulate silica sorbents are commonly employed materials for use as stationary phases in reversed-phase liquid chromatography (RPLC). Monomeric octadecyl (C18) modified silica prepared by reaction of monofunctional silanes with surface silanols is the most popular chromatographic sorbent used in RPLC and has been employed in the analytical separations of a wealth of solutes. “Shape-selective” polymeric C18 stationary phases prepared with solution-polymerized alkylsilanes are generally preferred over monomeric materials to improve separations of shape isomers, such as polycyclic aromatic hydrocarbons (PAH), polychlorinated biphenyl (PCB) congeners, steroids, and * Corresponding author. E-mail:
[email protected]. Telephone: (301) 9753116. Fax: (301) 977-0685. † Analytical Chemistry Division. ‡ Physical and Chemical Properties Division.
7862 Analytical Chemistry, Vol. 77, No. 24, December 15, 2005
carotenoids.1,2 Because the various isomeric forms of these compounds are known to have varying biochemical properties,3-5 separation of the individual species is necessary to assess potential health effects. Although the separation of the nonpolar isomers of carotenoids (e.g., cis and trans forms of β-carotene) is improved on shapeselective polymeric C18 columns,6-8 more selective separations result from a polymeric phase with an extended chain length (C30).8 This extended-length C30 phase was designed to increase the interaction of relatively large-size carotenoid molecules with the stationary phase and has been demonstrated to improve the separations of carotenes and xanthophylls,8-10 as well as retinoic acid,11 tocopherol,12,13 and vitamin K14 isomers. Investigations with other shape-constrained solutes (i.e., PAHs) indicated similar trends of enhanced shape selectivity for longer chain length alkyl stationary phases of both monomeric and polymeric types.15,16 In these chromatographic studies, SRM 869a Column Selectivity Test Mixture17 was used to determine the capacity factors (k′) for two PAH isomers of differing molecular shape, nonplanar tetrabenzonaphthalene (TBN) and planar benzo(1) Sander, L. C.; Pursch, M.; Wise, S. A. Anal. Chem. 1999, 71, 4821-4830. (2) Wise, S. A.; Sander, L. C. In Chromatographic Separations Based on Molecular Recognition; Jinno, K., Ed.; Wiley-VCH: New York, 1997; pp 1-64. (3) Sweeney, J. P.; Marsh, A. C. J. Nutr. 1973, 103, 20-25. (4) McFarland, V. A.; Clarke, J. U. Environ. Health Perspect. 1989, 81, 225239. (5) O’Neil, C. A.; Schwartz, S. J. J. Chromatogr. 1992, 624, 235-252. (6) Lesellier, E.; Tchapla, A.; Krstulovic, A. M. J. Chromatogr. 1993, 645, 2939. (7) Epler, K. S.; Sander, L. C.; Ziegler, R. G.; Wise, S. A.; Craft, N. E. J. Chromatogr. 1992, 595, 89-101. (8) Sander, L. C.; Sharpless, K. E.; Craft, N. E.; Wise, S. A. Anal. Chem. 1994, 66, 1667-1674. (9) Bell, C. M.; Sander, L. C.; Fetzer, J. C.; Wise, S. A. J. Chromatogr., A 1996, 753, 37-45. (10) Van Heukelem, L.; Thomas, C. S. J. Chromatogr., A 2001, 910, 31-49. (11) Strohschein, S.; Schlotterbeck, G.; Richter, J.; Pursch, M.; Tseng, L. H.; Ha¨ndel, H.; Albert, K. J. Chromatogr., A 1997, 765, 207-214. (12) Henry, C. W.; Fortier, C. A.; Warner, I. M. Anal. Chem. 2001, 73, 60776082. (13) Strohschein, S.; Pursch, M.; Lubda, D.; Albert, K. Anal. Chem. 1998, 70, 13-18. (14) Cook, K. K.; Mitchell, G. V.; Grundel, E.; Rader, J. I. Food Chem. 1999, 67, 79-88. (15) Rimmer, C. A.; Sander, L. C.; Wise, S. A.; Dorsey, J. G. J. Chromatogr., A 2003, 1007, 11-20. (16) Sander, L. C.; Wise, S. A. Anal. Chem. 1987, 59, 2309-2313. (17) Certificate of Analysis, Standard Reference Material 869a, Column Selectivity Test Mixture; National Institute of Standards and Technology, Gaithersburg, MD, 1998. 10.1021/ac051085v Not subject to U.S. Copyright. Publ. 2005 Am. Chem. Soc.
Published on Web 11/12/2005
[a]pyrene (BaP). The selectivity factor (RTBN/BaP ) k′TBN/k′BaP) was observed to decrease for an increase in alkyl chain length, which is indicative of an increase in stationary-phase shape recognition. Enhancement in shape selectivity has also been observed as a result of reduced column temperature.18-21 An investigation of chromatographic shape selectivity for both monomeric and polymeric C18 stationary phases over a wide range of temperatures (-17 to 99 °C) illustrates a steady increase in RTBN/BaP values with an increase in temperature.19 At any given temperature, the polymeric stationary phase exhibited enhanced shape recognition (lower RTBN/BaP values) compared to the less dense monomeric phase. The lack of sudden changes in RTBN/BaP values over this temperature range also suggests that these phases do not undergo abrupt structural changes (or phase transitions). In general, shape selectivity is influenced by any chromatographic parameter that increases alkyl chain conformation order. Spectroscopic studies22-32 of alkyl-modified surfaces have demonstrated that the conformational order of alkyl chains increases with both alkyl chain length and reduced temperature. Conformational order has also been observed to increase with alkyl chain density; the surface modification approach does not directly influence order.27,32,33 A recent review34 of chromatographic and spectroscopic studies provides a more complete summary of the factors that influence alkyl stationary-phase order and disorder in the context of shape recognition properties. A number of theoretical approaches through molecular simulation techniques have also been employed to elucidate the specific structural features that may control solute retention processes; these investigations are described in greater detail in part 1 of this series.35 Notably, these simulation studies depict monomeric-type RPLC phases up to a chain length of 18 carbons and thus are not appropriate for describing potential shape recognition processes on either shape-selective polymeric C18 or extended-length C30 alkylsilane stationary phases. (18) Hesselink, W.; Schiffer, R. H. N. A.; Kootstra, P. R. J. Chromatogr., A 1995, 697, 165-174. (19) Sander, L. C.; Wise, S. A. Anal. Chem. 1989, 61, 1749-1754. (20) Sander, L. C.; Wise, S. A. J. Sep. Sci. 2001, 24, 910-920. (21) Sentell, K. B.; Henderson, A. N. Anal. Chim. Acta 1991, 246, 139-149. (22) Albert, K. J. Sep. Sci. 2003, 26, 215-224. (23) Beaufils, J. P.; Hennion, M. C.; Rosset, R. Anal. Chem. 1985, 57, 25932596. (24) Cheng, J. L.; Fone, M.; Ellsworth, M. W. Solid State NMR 1996, 7, 135140. (25) Doyle, C. A.; Vickers, T. J.; Mann, C. K.; Dorsey, J. G. J. Chromatogr., A 1997, 779, 91-112. (26) Doyle, C. A.; Vickers, T. J.; Mann, C. K.; Dorsey, J. G. J. Chromatogr., A 2000, 877, 41-59. (27) Ducey, M. W.; Orendorff, C. J.; Pemberton, J. E.; Sander, L. C. Anal. Chem. 2002, 74, 5576-5584. (28) Jinno, K.; Wu, J.; Ichikawa, M.; Takata, I. Chromatographia 1993, 37, 627634. (29) Pursch, M.; Brindle, R.; Ellwanger, A.; Sander, L. C.; Bell, C. M.; Ha¨ndel, H.; Albert, K. Solid State NMR 1997, 9, 191-201. (30) Pursch, M.; Sander, L. C.; Egelhaaf, H. J.; Raitza, M.; Wise, S. A.; Oelkrug, D.; Albert, K. J. Am. Chem. Soc. 1999, 121, 3201-3213. (31) Sander, L. C.; Callis, J. B.; Field, L. R. Anal. Chem. 1983, 55, 1068-1075. (32) Singh, S.; Wegmann, J.; Albert, K.; Muller, K. J. Phys. Chem. B 2002, 106, 878-888. (33) Jinno, K.; Ibuki, T.; Tanaka, N.; Okamoto, M.; Fetzer, J. C.; Biggs, W. R.; Griffiths, P. R.; Olinger, J. M. J. Chromatogr. 1989, 461, 209-227. (34) Sander, L. C.; Lippa, K. A.; Wise, S. A. Anal. Bioanal. Chem. 2005, 382, 646-668. (35) Lippa, K. A.; Sander, L. C.; Mountain, R. D. Anal. Chem. 2005, 77, 78527861.(ac0510843).
In this work, we examine the influence of temperature and alkyl chain length on the order and disorder of the alkyl stationary phases through the use of molecular dynamic (MD) simulation techniques. Specifically, we have investigated the simulations of alkylsilane models of various chain lengths (C8, C18, C30) and selected surface coverages, together with two C18 models of monomeric and polymeric types that have typical surface coverages (3.28 and 4.91 µmol/m2, respectively) and that are subject to several temperature conditions in the range of -5 to 59 °C. These models have been designed to represent RPLC phases of varying chain lengths16 and column temperature conditions19,20 for which shape recognition capabilities were characterized by the shape-selectivity factor RTBN/BaP, providing a numerical assessment of how well the column will separate shape-constrained isomers. Values of RTBN/BaP < 1 are typical of phases with high (H) shape selectivity (e.g., polymeric C18 columns), values of RTBN/BaP between 1 and 1.7 are considered intermediate (I), and values of RTBN/BaP > 1.7 are more typical of columns with low (L) shape recognition.1 In the preceding paper in this journal,35 the effects of alkylsilane surface coverage and bonding chemistries on the conformation order of chromatographic models that represent both monomeric and polymeric C18 stationary phases were examined through comparable molecular dynamics simulation techniques. Alkyl chain order was observed to increase with increased surface coverage, and in general, monomeric and polymeric phases with similar surface coverages yield similar alkyl chain order. These results are consistent with experimental observations34 of order and disorder in RPLC stationary phases. These two parts of this simulation study represent the foundation of our shape-selectivity modeling efforts to investigate constrained solute retention processes. Fully solvated computational models of similar shapeselective RPLC systems will be the subjects of subsequent work. COMPUTATIONAL METHODS Chromatography Model Description. A silica model was constructed by slicing a 7.5-Å-thick segment of the (101) face of a quartz crystalline lattice with width dimensions of 41 × 59 Å. This surface cut results in a vicinal silanol density of 9.8 µmol/ m2, which is considerably greater than the commonly reported (8.0 ( 1.0 µmol/m2)36 silanol concentration of silica. The additional concentration of silanols was necessary to randomly distribute ligands on the surface over a wide range of densities. Monomeric- and polymeric-type alkylsilane ligands of carbon lengths 8, 18, and 30 were employed in these simulation models. The monomeric phases were constructed of dimethylalkyl ligands of various lengths (octyl, octadecyl, triacontyl) that were tethered to a silica surface via a single silanol. Trifunctional, trialkylsilane ligands of various lengths (originating from the solution reaction of the trichloroalkylsilane in water forming a polymer of 3 alkylsilane units) were selected to represent polymeric alkylsilane ligands and were also tethered to the quartz surface via one silanol. Chromatography models were created by covalently tethering alkylsilane ligands (with the alkyl chains in an initial all-trans conformation) to the silanol oxygen atoms of the quartz surface and at an orthogonal (⊥) position to the bulk quartz. Models of varying chain lengths and identical densities were created by (36) Nawrocki, J. J. Chromatogr., A 1997, 779, 29-71.
Analytical Chemistry, Vol. 77, No. 24, December 15, 2005
7863
Table 1. Structural Details of Simulated Alkylsilane Chromatographic Phase Models at Varying Temperatures
temp (°C)
temp (K)
average chain spacinga,b (Å)
-4.6 2.4 14.3 24.8
268.6 275.6 287.5 298.0
7.2 ( 0.7 7.2 ( 0.7 7.3 ( 0.7 7.2 ( 0.7
1.0 24.8 35.9 45.8 59.2
274.2 298.0 309.1 319.0 332.4
3.1 ( 0.2 3.1 ( 0.1 3.1 ( 0.2 3.1 ( 0.2 3.1 ( 0.2
C18 Trifunctional Polymeric (4.91 µmol/m2 Density) 22.0 ( 1.2 21.8 ( 1.3 6.6 ( 4.2 21.3 ( 1.3 20.9 ( 1.6 8.5 ( 5.0 21.2 ( 1.7 20.9 ( 1.9 9.6 ( 5.3 21.1 ( 1.8 20.3 ( 2.4 15.2 ( 7.5 19.9 ( 2.7 19.2 ( 3.2 13.9 ( 9.5
12.1 19.2 19.8 19.8 26.5
0.28 (H) 0.71 (H) 0.89 (H) 1.00 (I) 1.21 (I)
24.8 59.2
298.0 332.4
3.0 ( 0.2 3.0 ( 0.2
C30 Trifunctional Polymeric (4.30 µmol/m2 Density) 37.0 ( 1.1 36.8 ( 1.2 6.0 ( 2.8 36.6 ( 1.4 36.4 ( 1.5 4.2 ( 2.6
9.5 10.7
0.55 (H)
average end-to-end chain lengthb,c (Å)
average phase thicknessb,d (Å)
average tilt angleb,e (θ)
C18 Monomeric (3.28 µmol/m2 Density) 19.9 ( 2.2 18.4 ( 3.3 14.4 ( 8.9 19.1 ( 2.1 19.0 ( 2.6 16.6 ( 8.2 18.6 ( 3.1 16.8 ( 3.5 23.7 ( 10.9 16.8 ( 2.9 15.7 ( 3.3 19.7 ( 10.3
average % of gauche dihedrals
phase selectivity RTBN/BaPf and categoryg
25.0 26.4 29.5 38.4
0.79 (H) 1.00 (I) 1.43 (I) 1.72 (L)
a Distance between Si atoms of the alkylsilane chains. b Uncertainty represents standard deviation of 100 measurements. c Distance between Si atom and end C atom of the alkylsilane chain. d Magnitude of the distance vector between Si and end C of the alkylsilane chain perpendicular to the quartz surface. e Relative to the axis perpendicular to the quartz surface. f Experimentally determined selectivity coefficients for C18 RPLC phases at various temperatures that are represented by simulation models (see ref 19 for more details). g Shape-selectivity of stationary phases categorized as high (H; RTBN/BaP < 1), intermediate (I; 1 < RTBN/BaP < 1.7), and low (L; RTBN/BaP > 1.7).
extending the length of the existing alkyl chains (i.e., C18) without changing their placement on the silica surface. A 15-Å vacuum layer was fixed above the ligands, resulting in a unit cell height of 35, 50, and 65 Å for models with alkyl chain lengths of 8, 18, and 30, respectively. Structural representations of these chromatography models are illustrated for the C18 models in Figure 1 of paper 1 of this series.35 Molecular Dynamics. The molecular systems were subject to a constant-volume, constant-temperature MD simulation method using the Discover molecular dynamics module employing the COMPASS force field37,38 within Cerius2 (v. 4.6) and Materials Studio (v. 2.2) software (Accelrys, Inc., San Diego CA). An atombased summation method was utilized with the nonbonded interactions cutoff set to a distance of 11.0 Å, accompanied by a spline width of 1.0 Å and a buffer width of 0.5 Å. The conformation of the initial models with 3D periodic boundary conditions was first minimized using the steepest descent convergence method, followed by the conjugate gradient method until the convergence reached 0.1 kcal/mol‚Å. The constant-volume, constant-temperature systems were then subject to equilibration for 100 ps with a time step of 1 fs. Temperature control was maintained by the Nose´-Hoover thermostat39 at the appropriate temperature (269-332 K). MD simulation was then allowed to progress with the same operational conditions for up to 2000 ps. The atomic positions of the bulk quartz surface (with the exception of the exposed surface silanols) were held constant during both equilibration and simulation steps. The atomic coordinates of the MD simulation models were recorded at 1-ps intervals for analysis. Structural Analysis. Following MD simulation, selected structural parameters that describe the order of alkylsilane chains of the model were determined for each picosecond over a total of 100 ps. These included the alkylsilane chain spacing and length, (37) Sun, H.; Rigby, D. Spectrochim. Acta, Part A 1997, 53, 1301-1323. (38) Sun, H. J. Phys. Chem. B 1998, 102, 7338-7364. (39) Hoover, W. G. Phys. Rev. A 1985, 31, 1695-1697.
7864 Analytical Chemistry, Vol. 77, No. 24, December 15, 2005
phase thickness, chain tilt angle, and dihedral angles as well as the carbon density profiles in the direction orthogonal to the quartz surface (z) and are described and illustrated in further detail in part 1 of this series.35 The molecular-level features within each model (i.e., chain length, tilt angle (θ)) were averaged for each alkyl chain over a 100-ps interval; the model was deemed stable when the 100-ps interval averages of these parameters were constant over four successive intervals (400 ps total). These parameters remained constant by 1600 ps for the bulk of the models. RESULTS AND DISCUSSION A group of alkylsilane chromatography models of various chain length (C8, C18, C30) of both monomeric and polymeric types were constructed over a range of densities (2.52-5.94 µmol/m2) and subject to various temperatures (-4.6 to 59.2 °C) under molecular dynamic simulation conditions. The structural details of two C18 models that represent monomeric and polymeric stationary phases at typical surface coverages (3.28 and 4.91 µmol/m2, respectively) and subject to various temperature conditions19 are presented in Table 1. Chromatography models of chain lengths C8, C18, and C30 were also constructed at room temperature (24.8 °C) to represent RPLC stationary phases of differing chain lengths that also have been experimentally characterized for their shapeselectivity capacity;16 these data are compiled in Table 2. The shape-selectivity numerical descriptor (RTBN/BaP) and the categorical ranking of the relevant RPLC phase that corresponds to each model are denoted in both Tables 1 and 2. Side- and top-view snapshots of monomeric and polymeric C18 chromatographic phase models over the range of temperatures examined are provided in Figure 1 and represent the atomic positions of the last measurement within the 100-ps time frame used to collect the structural data. Model snapshots for selected monomeric and polymeric chromatographic phase models of various carbon lengths are illustrated in Figure 2 and are labeled with the corresponding shape-selectivity numerical descriptor
Table 2. Structural Details of Simulated Alkylsilane Chromatographic Phase Models of Varying Chain Lengths (24.8 °C) average chain spacingb,c (Å)
average end-to-end chain lengthc,d (Å)
average phase thicknessc,e (Å)
average tilt anglec,f (θ)
average % of gauche dihedrals
phase selectivity RTBN/BaPg and categoryh
phase typea
ligand type
chain density (µmol/m2)
m p
monofunctional C8 trifunctional, tri-C8
2.52 6.76
7.3 ( 0.6 3.0 ( 0.2
C8 Chain Length 7.8 ( 1.0 9.6 ( 0.7
6.7 ( 0.7 9.1 ( 1.2
26.3 ( 13.4 13.5 ( 9.2
43.6 25.5
1.73 (L) 1.72 (L)
m p p p p
monofunctional C18 difunctional, tri-C18 trifunctional, tri-C18 trifunctional, tri-C18 trifunctional, tri-C18
3.28 3.89 4.30 4.91 5.94
7.2 ( 0.7 3.1 ( 0.1 3.1 ( 0.2 3.1 ( 0.1 3.0 ( 0.2
C18 Chain Length 16.8 ( 2.9 21.2 ( 1.9 21.3 ( 1.9 21.3 ( 1.3 22.2 ( 0.9
15.7 ( 3.3 20.2 ( 5.0 21.1 ( 2.5 20.9 ( 1.6 22.1 ( 1.0
19.7 ( 10.3 13.7 ( 20.4 9.3 ( 7.4 8.5 ( 5.0 6.1 ( 2.7
38.4 18.7 18.2 19.2 10.4
1.72 (L) 1.61 (I) 1.11 (I) 0.71 (H) 0.49 (H)
m m p p p
monofunctional C30 monofunctional C30 trifunctional, tri-C30 difunctional, tri-C30 trifunctional, tri-C30
2.73 4.91 4.30 4.91 4.91
7.3 ( 0.6 5.2 ( 0.9 3.0 ( 0.2 3.1 ( 0.2 3.0 ( 0.2
C30 Chain Length 34.1 ( 1.9 36.3 ( 1.0 37.0 ( 1.1 36.8 ( 1.1 37.3 ( 0.9
33.4 ( 2.2 35.9 ( 1.0 36.8 ( 1.2 36.8 ( 1.4 37.0 ( 1.0
11.8 ( 4.2 8.0 ( 2.3 6.0 ( 2.8 4.3 ( 2.5 6.9 ( 2.1
16.0 10.7 9.5 9.7 7.6
0.55 (H) 0.55 (H)
a m, monomeric; p, polymeric phase type. b Distance between Si atoms of the alkylsilane chains. c Uncertainty represents standard deviation of 100 measurements. d Distance between Si atom and end C atom of the alkylsilane chain. e Magnitude of the distance vector between Si and end C of the alkylsilane chain perpendicular to the quartz surface. f Relative to the axis perpendicular to the quartz surface. g Experimentally determined selectivity coefficients for RPLC phases at room temperature and various chain lengths that are represented by simulation models (see ref 16 for more details). h Shape selectivity of stationary phases categorized as high (H; RTBN/BaP < 1), intermediate (I; 1 < RTBN/BaP > 1.7), and low (L; RTBN/BaP > 1.7).
RTBN/BaP. These snapshots are intended to illustrate the structural trends within the computational models and are to be used to complement, but not substitute for, the quantitative analysis of the structural parameters calculated for the models. Chain Gauche Defects. Generalized trends of the effect of temperature on overall chain order can be made through an examination of the average percentage of gauche dihedrals (defined as a torsion angle magnitude greater than 66°) for the monomeric and polymeric C18 stationary-phase models (Table 1). A steady decrease from 38 to 25% in average chain gauche dihedrals for the 3.28 µmol/m2 monomeric C18 model is observed as the temperature of the simulation is decreased from 24.8 to -4.6 °C. This trend is consistent with the review24,26,31-34,40 of spectroscopic (Fourier transform infrared (FT-IR), Raman, nuclear magnetic resonance (NMR)) evidence for which a gradual increase in alkyl chain order in comparable RPLC phases was observed through various spectroscopic indicators for a decrease in temperature, with no significant indication of phase transition behavior. Only in the case of octadecylsilane phases with low surface coverage (1.6 µmol/m2) was a slight phase transition region at 20 °C observed to occur.41 Previous molecular dynamic simulation studies of monomerictype C18 models also illustrate similar temperature effects that are consistent with these results. Ban et al.42 observed a decrease in the fraction of gauche conformations for a decrease in temperature for a monomeric-type C18 model with a surface coverage of 1.9 µmol/m2. Beck and Klatte43 observed roughly 25% gauche defects in their simulations of monomeric-type C18 models over a range of surface coverages at room temperature but also noted that (40) Ducey, M. W.; Orendorff, C. J.; Pemberton, J. E.; Sander, L. C. Anal. Chem. 2002, 74, 5576-5584. (41) Ho, M.; Pemberton, J. E. Anal. Chem. 1998, 70, 4915-4920. (42) Ban, K.; Saito, Y.; Jinno, K. Anal. Sci. 2004, 20, 1403-1408. (43) Klatte, S. J.; Beck, T. L. J. Phys. Chem. 1995, 99, 16024-16032.
similar values were observed at lower temperatures. An increase in temperature was observed to have a greater influence on the diffusion constants of the tail carbons rather than on the atoms in the middle of the chain. Overall, no phase transition was indicated as the diffusion constants were observed to gradually increase over the temperature range of 200-300 K. A similar reduction in gauche dihedrals was also observed for the 4.91 µmol/m2 polymeric C18 model, with a steady decrease from 26 to 12% but over a broader temperature range (59 to 1 °C). Note that, at room temperature, this polymeric model has considerable less gauche defects (19%) present in the chains than the monomeric model (38%). This is attributable to the significant difference in alkyl surface coverage between the more dense polymeric model compared to the less dense (3.28 µmol/m2) monomeric model and is described in more detail in the paper preceding this work. A recent FT-IR study of a polymeric n-alkylmodified silica gel32 at a surface coverage of 4.7 µmol/m2 also estimated that ∼20% of C18 chain dihedrals were gauche conformers at room temperature; this value was only nominally reduced over a temperature range of 250-300 K (-23 to 27 °C) applicable to this study, but was more significantly reduced at lower temperatures (
