Interest of HPTLC for hydrocarbon-related analysis and possibilities of detection by induced-

fluorescence

Vicente L. Cebolla

CSICInstituto de Carboquímica

(Zaragoza, Spain)

Detection of molecules by fluorescenceNative fluorescence

Use of a derivatizing agent

Fluorescence only occurs in certain molecules

Can all the non-fluorescent molecules be determined by fluorescence?

SATURATED HYDROCARBONS(Alkanes, isoalkanes and cycloalkanes)

They are in high concentration in petroleum and its derived products

They have traditionally been considered as inert molecules

They have neither UV nor fluorescence spectra under analyticalworking conditions

What is Fluorescence Detection by Intensity Changes (FDIC)?

A wide number of compounds, including non-fluorescent ones, induce changes in the fluorescence spectra (either increase or quenching) of certain probes that exclusively affect emission intensity

Probes:

Berberine cation

OO

NH3CO

OCH3

N

OCH3OCH3

H3CO

H3COCH3

Coralyne cation

NO

Reichardt's Dye

J.Chromatogr.A 2007, 1146, 251Anal.Chem. 2006, 78, 3699Anal.Chem. 2000, 72, 1759Org.Lett. 2000, 2, 2311

FDICHPLC detection

HPTLC detection

Theory

Sensor films

FDIC produced by saturated hydrocarbons (alkanes) in HPTLCIncrease of fluorescent emission

+hν

N

OCH3OCH3

H3CO

H3COCH3

* Fluorescent response depends on alkane mass and chain length, and probe concentration

Number of C10.0 15.0 20.0 25.0

Are

a co

un

ts

0500

1000150020002500300035004000

Coralyne (6 ppm)(λexc= 410 nm)

Berberine (60 ppm)(λexc= 365 nm)C12

C16

C18

C22C24

Elution lenght (mm)

Berberine 60 ppm

0

0.5

1.0

1.5

2.0

2.5

120 140 160

2 μg

6 μg4 μg

2 μg

8 μg4 μg

12 μg

C24

C16120 140 160

berberine

Fluo

resc

ent

Resp

onse λexc= 365 nm

C18 C16

TLC HPTLC

λem> 450 nm

Separation and detection of lipid mixtures

Fluo

resc

ent r

espo

nse

L-α-

phos

phat

idile

than

olam

ine

L-α-

phos

phat

idili

nosi

tol

L-α-

phos

phat

idilc

holin

e

L-α-

lisop

hosp

hatid

ilcho

line

Elution length (mm)

Elution length (mm)30 40 50 60 70 80 90

Flu

ores

cen

ce r

esp

onse

(a

-0.20

0.20

0.60

1.00

1.40

1.80

2.20

trio

lein

Ole

ic a

cid

met

hyl e

ster

Ole

ic a

cid

Cho

lest

eryl

olea

te

Cho

lest

erol

NEUTRAL LIPIDS (HPTLC)Coralyne (6 ppm). PreimpregnationPetroleum ether (80)-diethylether (20)-acetic acid (1), 20 minλexc= 410 nm; λem> 450 nm

PHOSPHOLIPIDS (in TLC plates)Berberine (10 ppm).PostimpregnationChloroform (65)-methanol (25)-water(4), 15 minλexc= 365 nm; λem> 450 nm

HPTLC-FDIC application to lipids

A polar antibiotic

Elution length (mm)45.0 50.0 55.0 60.0 65.0 70.0 75.0 80.0

Res

pons

e (a

.u.)

-2.50

-2.00

-1.50

-1.00

-0.50

0.00

0.50

1.00

λexc= 410 nm

UV at 254 nm

3 μg

1 μg

1 μg

3 μg

HPTLC layer, methanol, 5 min; λem> 450 nm

Coraline6 ppm

Other compounds give quenching of emission

An aminoacid

HN

NH+H2C

HOOC

CH2N

H

Application point

UV at 254 nm

Λexc= 365 nm

Histidine

OHOH

OH

NH

OO

HO

H3CO

O

OO

O

NN

N

Rifampicine

Berberine60 ppm

Mecanismo en capa finaMecanismo en capa finaMecanismo en capa finaMecanismo en capa fina

Electrostatic interactions are responsible of emission increases (MODEL)

Two effects, through polarizability α

nrr

r

k kk+

=φVaría n, a través de 1) εr, n, α

2) The higher the interaction, the higher the protection(the higher α)

A theoretical model to explain changes in emission

MODEL: non-specific, dipolar interactions between probe and the corresponding analytecontribute to the efficiency of the fluorescence emission, creating a microenvironment that isolates the fluorescent probe and prevents non-fluorescent decay mechanisms.

FDIC response depends on the balance non-specific (electrostatic) and specific interactions between probe-analyte, as well as the dielectric permittivity (εr) of the medium. It only involves non-covalent interactions

Negative peaks can be attributed to quenching produced by net specific interactionsbetween polar compounds and probe

Problems in hydrocarbon analysis: heavy boiling complex mixturesHydrocarbon Type Analysis

Almost all types of petroleum products can be analyzedusing a single techniqueDirect and accurate determination of saturated compoundsusing FDIC

TLC / HPTLC

DETERMINATION: column techniques are not satisfactory

TLC is adequate to the analysis of complex mixtures* Analysis of the whole sample* Uneluted compounds are also detected* Scanning of the same sample under different conditions

Conventional TLC system:

TLC plates (5-25 μm particle size)Autospotter for sample applicationConventional vertical tank for elution

HPTLC system:

HPTLC plates (3-10 μm particlesize)

Linomat for sample applicationHorizontal developing chamber

Experimental systems used in our research

Conventional TLC for hydrocarbon analysis

Hydrocarbon-Type Analysis of fossil fuel-derived products

Two measurements:

- Saturates developed on a berberine-impregnated layer and detected by fluorescence; λexc=365 nm; λem> 450 nm- The other peaks on nonimpregnated layer and detected by UV at 254 nm

Analyte load and structure

Chromatographic variables

Equipment variables- Development system - Application system

- Probe nature - Probe concentration- Application volume- Scanning beam size- Excitation and emission - Development length

1.6

1.2

0.8

0.4

0

2

1.5

1

0.5

0

0 50 100 150 200

1

0.8

0.6

0.4

0.2

0

2

1.5

1

0.5

0

100 200120 140 160 180

Elution length

Heavy oil

Fuel de vis-breaking

Lubricant

Satu

rate

sS

atur

ates

Sat

urat

esS

atur

ates

Aro

mat

ics

Aro

mat

ics

Aro

mat

ics

Aro

mat

ics

Pol

ars

Pola

rs

Une

lute

d

Heavy Oil Lubricant

Visbreaking fuel Gas Oil

Results using conventional TLC

Percentages of saturates are in agreement with those obtainedusing the reference techniques

Sample wt % ref. % RSD Ref.tech. int. (μg) r2

Heavy oil 35,2 36,1 6,85 (TLC-FID) 3-16 0,999Visbreaking fuel 12,3 12,1 1,85 (TLC-FID) 4-20 0,996Lubricant 56,2 56,8 2,21 (ASTM D2007) 4-16 0,993Gasoil 69,8 70,8 (HPLC-RI) 5-20 0,99

SATURATES

HPTLC for hydrocarbon analysis

GAS OIL

Alkanes Naphtenes Total aromatics

Heavy PACs

silica gel impregnated with berberine

- 5 minutes with n-hexane

Fluorescence(FDIC)

silica gel

- 5 minutes with n-hexane- 3 minutes with acetone

UV at 254 nm

silica gel impregnated with caffeine

- 5 minutes with n-hexane- 1 minute with acetone

Fluorescence(native)

scanning densitometry detection

Hydrocarbon-type analysis of a heavy gasoil

1. Alkanes by FDIC

25 20 15 10 5 0

0

0.5

1.0

1.5

2.0

2.5 ALKANES

NAPHTHENES

AROMATICS

detection by UVdetection by fluorescence

berberineCalibration Range : 0.05 to 1.5 µg y=1163x+744 r2 = 0.9862LOD = 0.05 µg LOQ = 0.15 µg

Detection by UV

Detection by FDICλexc = 365 nmλem = > 450 nm

- developement: 5 minutes with n-hexane

Is it possible to determine all families by FDIC?

Either on the same impregnated plateor on a non-impregnated plate

2.- Naphthenes by FDIC

65 55 45 35 25 15Distance of migration (mm)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7Response (AU)

210 nm

NAPHTHENES absorb at 210 nm

ALKANES

λexc = 365 nmλem = > 450 nm

Calibration Range : 0.6 to 2.4 μg y=655x-89 R2 = 0.9097 LOD = 0.1 μg LOQ = 0.3 μg

min10 20 30 40 50 60 70 80 90 100

counts

5000

10000

15000

20000

GCGC--FIDFID

11HH--NMRNMR

Chemical shifts and coupling constant typical

of cyclic -CH2-

IRFTIRFT

Increasing of the ratio CH2/CH3

(1459 cm-1 - 1380 cm-1)

Effect of berberineon separation?Charge-transfer?

3.- Total aromatics by UV

UV Scanning Densitometry

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

UV Response (AU)

Distance of migration (mm)0510152025

After the 2nd migration step

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

UV Response (AU)

Distance of migration (mm)0510152025

Acetone Calibration Range : 0.1 to 2.0 µg y=873x+89 r2 = 0.9684 LOD = 0.06 µg LOQ = 0.18 µg

1115 7

BaP

PerCor

On a non-impregnated silica gel HPTLC plate because berberine is highly

soluble in acetone

HPTLC for aromatics (PACs)

Influence of caffeine in separation of aromatics

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Response

45 35 25 15 5

Distance of migration (mm)

Silica gel

0.0

0.2

0.4

0.6

0.8

1.0

Response

45 35 25 15 5

Distance of migration (mm)

UV 300nm

NativeFluorescence

365 nm

Caffeine stationary

phase

Plate impregnated with Caffeine : electron acceptor

Selective detection and quantification of HEAVY PACs

Injection point

55 55 45 35 25 15 5Distance of migration (mm)

0.00

0.05

0.10

0.15

0.20Response (AU)

S

S +

UV at 300 nm

Response

0.0

0.2

0.4

0.6

0.8

1.0

45 35 25 15 5

Distance of migration (mm)

NativeFluorescence

2nd migration step (EtOH or acetone) to merge PACsCalibration Range : 0.1 to 2.0 μg y=898x+8 R2 = 0.9970 LOD = 0.07 μg LOQ = 0.25 μg

λexc = 365 nmλem = > 450 nm

A B

C

A: 2 aromatic rings and benzothiophenesB: 3 aromatic rings + fluorenic+ dibenzothiophenicC: four and more aromatic rings

4

5 andmore

Separation of aromatics according to ring number

0.0

0.2

0.4

0.6

Response

45 35 25 15 5

Distance of migration (mm)

I-II rings

III rings

IV and+IV rings

80 60 40 20 0

0.05

0.15

0.25

0.35

Distance of migration (mm)

Response

Paper Bridge

Solvent SolventSolvent

caffeine

Silica gel

Hexanecaffeine

Silica gel

DCM

…and AMD?

Calibration in Hydrocarbon-Type Analysis

Calibration: which standards?

One molecule

Complex mixtures

Its commercial standard

One molecule as representativestandard ?

Preparative step based on TLC

Response factor is generally structure dependent

Detection method is chosen regarding to the structure of molecules

Possibilities:Internal standardA representative alkane (SIMDIS concept)A representative fraction

Standard purification

Preparative TLC

Mig

rati

on

Soxtecextraction

Extract concentrated under N2 until

constant weight

Purity checked

GC-MSTLCNMR

Each fraction will be a standardEach fraction will be a standard

25 20 15 10 5 0

0

0.5

1.0

1.5

2.0

2.5 ALKANES

NAPHTHENES

AROMATICS

detection by UVdetection by fluorescence

GC-FID of gas oil

GC-FID of alkanesfraction

Evaluation of evaporation losses during purification

6 % for alkanes 1 % for aromatics

Percentage (m/m)

determined by standard

addition

Percentage (m/m)

determined by external

standard (RSD%)

LCO 57.7 % 57.5 % (4.3)

GO 16.1 % 17.0 % ( 3.9 )

Standard Addition External Standard

Preparation of Synthetic Fuel

Theoretical value (%) Measured value (%)

(RSD%)

Recovered value (%)

16.8 15.3 (6.5) 91.1

22.2 22.7 (20.0) 102 . 2

32.2 29.8 (10.0) 92.5

52.3 47.1 (4.1) 90.1

Accuracy

HPTLC-FDIC application to other hydrocarbon-related samples

L-PP(30 μg)

Polar-aromatics

Polars

Fluo

resc

entr

espo

nse B-PP

(30 μg)

Saturates

Saturates

n-tetracosane, C24

(4 μg)

Elution distance (mm)

App

licat

ion

poin

t

Polar-aromatics

Polars

Liquids from lignine-polypropylene copyrolysis

Liquids from beech-polypropylene copyrolysis

Merci de votre attention

Research team on FDIC phenomena

University of Zaragoza:

CSIC, Instituto de Carboquímica:Luis Membrado, Eva Gálvez, Arancha Delgado

Jesús Vela, Javier Galbán, Rosa Garriga, Elena Mateos, Susana de Marcos,Andrés Domínguez, Isabel SanzUniversity of the Basque Country at San Sebastián:Fernando CossíoUniversité Paul Verlaine-Metz (France):Muriel Matt