New ways to characterize 'hard' and 'soft' surface properties by IGC-ID. Eric BRENDLE

Transcription

1 New ways to characterize 'hard' and 'soft' surface properties by IGC-ID Eric BRENDLE

2 Content IGC at Infinite Dilution Principle and Assumptions Real surfaces and accessibility Morphology index (IM) and applications Real surfaces with Soft material 4Molecular mobility 4Relative morphology index (RIM) Application of RIM to Soft and Hard surfaces Conclusions 2

3 IGC at Infinite Dilution Most used IGC technique Mainly for the determination of the surface energy (γ s d ) γ s d determination method: Injection of an homologous set of alkanes probes (3 or more) Plot G a versus Carbon Number 15 Nonane G a [kj/mol] 10 5 Hexane Heptane Octane Slope [ Ga(CH 2 )] Carbon Number or X T

4 IGC-ID on silica Comparison of two silica Pyrogenic Silica (Aerosil 130) Crystalline and layered Silica (H-magadiite) Same chemical composition, but different structures Measurements at 110 C Huge difference in terms of γ sd values! Ga [kj/mol] γ d s = mj/m2 γ d s = 40.4 mj/m2 Why such differences? What does it mean? 0 Aerosil 130 H-magadiite Carbon Number or X T

5 IGC-ID assumptions Dorris and Gray Probes visit all the surface Low probe amounts are injected to avoids the probe-probe interactions Probes adsorb flat on the surface Only adsorption occurs, i.e. 2D interactions (no absorption, i.e. 3D interactions) The surface is homogeneous Topologically (flat) Chemically Also valid for contact angle measurements Real surfaces are seldom flat and homogeneous! How assess if the surface is flat?

6 Real surfaces and accessibility n-octane isooctane cyclooctane Depending on the molecular structure, the accessibility to the surface, i.e. the interaction energy, will no be identical Comparing the behaviour of linear, branched and cyclic alkanes is able to deliver information about the surface nanoroughness How to describe the molecules? (i.e. taking into account their structure)

7 χ T : Molecular descriptor It is the oldest topological index related to molecular branching (1947) W is representative of the van der Waals volume of the molecule Therefore representative of their dispersive interaction ability W takes into account the structure of the molecules χ T based on the Wiener Index (W) [X T =f(w)] f(w) was developed so that χ T = Nc (carbon number) for n-alkanes XT or carbon number 10 5 butane propane 0 nonane octane heptane hexane pentane 2,3-dimethylpentane ethane methane Wiener Index decane For 2,3-dimethylpentane W=46

8 On Aerosil 130 Branched and cyclic probes are on the n-alkanes straight line All the probes have the same accessibility to the surface Aerosil 130 is flat at the molecular level A Ga [kj/mol] Ga(CH 2 )= 2.55 kj/mol Carbon Number or X T n-alkanes cycloalkanes methyl alkanes dimethyl alkanes trimethyl alkanes

9 On H-magadiite Branched and cyclic probes are significantly below on the n-alkanes straight line All the probes have not the same accessibility to the surface Strong size exclusion effects exist The most excluded are the cyclic probes The less excluded are the 2-methylalkanes Ga [kj/mol] n-alkanes cycloalkanes methyl alkanes dimethyl alkanes trimethyl alkanes Ga(CH 2 )= 6.41 kj/mol = mj/m2 γ d s 0 Ga(CH 2 )= 3.14 kj/mol γ d s = 60.1 mj/m Carbon Number or X T How to characterize the size exclusion effects?

10 IM: Morphology Index 2-methylpentane: χ T =5,84 Ga [kj/mol] Measured t N =0.84 min n-alkanes cycloalkanes methyl alkanes dimethyl alkanes trimethyl alkanes Expected if same accessibility as n-alkanes t N =7.44 min Ga(CH 2 )= 6.41 kj/mol = mj/m2 γ d s Ga(CH 2 )= 3.14 kj/mol γ d s = 60.1 mj/m2 Assuming the same behaviour than n-alkanes For 2-methylpentane: IM=0.11 (i.e. 11% of the expectedt N ) For flat surfaces IM Carbon Number or X T IM(2-MeC n ) > IM(2,2-diMeC n ) > IM(triMeC n ) > IM(cycloalkanes)

11 Pyrogenic Silica (amorphous) Comparison of samples issued from several manufacturers 50 < S BET < 400 m 2 /g Clear tendencies γ s d values increase with S BET γ s d values related to the nanoroughness (IM(cyclooctane)) Microporosity strongly influences the measured γ s d value γ d s [mj/m2 ] γ d s [mj/m2 ] S BET [m 2 /g] HDK microporous HDK Aerosil Cabosil HDK microporous HDK Aerosil Cabosil IM(cyclooctane) increasing nanoroughness 0,2 0,4 0,6 0,8 1,0

12 Partial conclusions for hard and clean surfaces A lot of surface structures provide adsorption sites where the n-alkanes interact with several surfaces at the same time Lead to stronger interactions The γ sd values are greatly influenced by the surface topology (nanoroughness) Nanoroughness includes: porosity, microporosity, defects, specific structures The use of branched and cyclic alkanes are efficient to describe the surface nanoroughness IM(cyclooctane) IM(isooctane) 1 What happens if the surface is organically modified? - Grafted solid - Coated surfaces - Polluted surfaces

13 Silica grafting Octyltriethoxysilane grafting on SiO 2 (Ultrasil 7000 GR), S BET 175 m 2 /g Comparison of the γ s d value evolution according to the grafting ratio Significant decrease of γ s d On the starting sample, cyclooctane is below the reference straight line On the most grafted sample, cyclooctane is above that line 110 C 25 γ d s = 64.8 mj/m γ d s [mj/m2 ] Ga [kj/mol] γ d s = 31.6 mj/m ,2 0,4 0,6 0,8 Grafting Ratio [mg/m 2 ] X T

14 IM [n.u.] 1,2 1,1 1,0 0,9 0,8 Different perception isooctane cyclooctane Grafted 110 C Sized Carbon Fiber@ 30 C IM [n.u.] 3,0 2,5 2,0 1,5 1,0 isooctane cyclooctane 0, Coverage [% in weight] Comparison of the evolution of the IM according the grafting ratio IM(isooctane) remain constant IM(cyclooctane) increases according the grafting ratio (even >1 for the highest ratio) IM(cyclooctane)>IM(isooctane) with increasing grafting ratio Also observed for coated surfaces and polymers near their T g Sometimes IM(cyclooctane)>2.5 (i.e. t N is 2.5 times higher as expected according n-alkanes) 0,5 0 0,2 0,4 0,6 0,8 1,0 1,2 Coverage [% in weight] Why? The solubility parameters of the n-alkanes, branched alkanes and cyclic alkanes explain their different behaviours δ cyclic >δ n-alkanes > δ branched

15 The RIM index Cases Surfaces Molecular Mobility IM behaviour Liquids, molten polymers Highly coated surfaces Clean solids (crystalline or not) Mineral amorphous solids Soft High (3D) IM(isooctane) < 1 < IM(cyclooctane) Hard None (2D) IM(cyclooctane) < IM(isooctane) 1 Polluted solid surface Partly coated surfaces Organic amorphous solids Hard + Soft or amorphous + crystalline Mixed IM(isooctane) < IM(cyclooctane) 1 Definition of the RIM index: 1.2 Mix: cohabitation soft and hard surfaces New representation: RIM vs IM(cyclooctane) RIM [n.u.] Clean and nanorough surfaces Soft surfaces (3D) IM(cyclooctane) [n.u.]

16 RIM: Application to the previous examples A clear picture of the different behaviours is obtained The effect of microporosity (Carbon Black, Graphites, microporous HDK) The effect of slight nanoroughness (pyrogenic silica) The presence of soft material (grafted or coated) RIM [n.u.] 2,5 2,0 1,5 1,0 0,5 0 Mix: soft and hard Clean and nanorough 0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 Soft surfaces (3D) Ultrasil HDK HDK microporous Sized Fibres Carbon Black Graphites Graphitized CB IM(cyclooctane) [n.u.]

17 RIM: Application to Lactose Easy distinction between the lactose samples From one provider to the other Sample showing a divergent behaviour The presence of soft material (API) RIM [n.u.] 1,2 1,0 0,8 Mix: soft and hard surfaces TAPI FAPI Soft surfaces (3D) Tablettose Fastflow Pharmatose API Lac+API (6%) Blind test 0,6 Clean and nanorough surfaces 0,5 0,6 0,7 0,8 0,9 1,0 1,1 1,2 IM(cyclooctane) [n.u.]

18 Conclusions On hard and clean surfaces The morphological index is able to describe the surface nanoroughness IM(cyclooctane) IM(isooctane) 1 The γ sd values are significantly impacted by the surface nanoroughness On soft or organically modified surfaces The RIM index is able to detect and describe the influence of molecular mobility The cyclic probes interact more strongly with the soft materials (dissolution) Opens the way to promising applications: Characterisation of the surface cleanliness Estimation of the amorphous contains (ex: lactose)

19 Perspectives Neuronic: a solution for automated IGC measurements rtners eimer Str. 55A, eim am Taunus, ckriem@porotec.de Simultaneous and independent measurements on two samples (2 analytical channels) Up to 30 test solvents available (up to 45) Working temperature ( C) Reduced dead-volume for high efficiency Bahnhofstrasse Sisseln Switzerland Real IGC-ID measurements Tel: ralf.duempelmann@inolytix.com γsd, nanoroughness (IM & RIM), acid-base, diffusion, HSP IGC-FC measurements Desorption isotherm, specific surface area, surface heterogeneity (AEDF) and thermal desorption ic BRENDLE. rendle@adscientis.com Example of performance ject Manager ean-marc ROTH th@adscientis.com tler heim Business enquires HSP determination on two samples (liquids) using 20 test solvents within a working day Mrs Michèle MABIN m.mabin@adscientis.com (repeatability tests included) Tél. : Fax :