application handbook fast gc/gcms · 2018. 11. 1. · application handbook fast gc/gcms...
TRANSCRIPT
Application Handbook
Fast GC/GCMSChromatography Volume 2
Fast GC and GC/MS using narrow-bore columns: Principles and Applications
Prof. Luigi Mondello, University of Messina, ItalyDr. Hans-Ulrich Baier, Shimadzu Europa GmbH, Germany
Application BookFast GC/GCMS
I.
II.
III.
IV.
V.
VI.
VII.
VIII.
IX.
Content
Overview 7
Basics of separation in fast GC 111. Column and stationary phase2. Length and peak resolution3. Inner diameter and peak resolution4. Film thickness and peak resolution5. Type of carrier gas and average linear velocity6. GC hardware requirements
Injection requirements in fast GC 15
Detector requirements in fast GC 191. Conventional detectors2. Mass spectrometric detector (quadrupole MS)
Applications in food analysis 231. Organophosphorous pesticides in food matrices
(GC-FPD, GC-FTD, GC/MS)2. The analysis of organochlorine pesticides in food matrices
using GC-ECD and NCI GC/MS. 3. Analysis of butter fatty acid methyl esters (GC-FID)
Application in the petrochemical field 31Kerosene analysis (GC-FID)
Applications in the fields of flavors & fragrances 321. Potential allergens in perfumes 2. Quality control of flavors (GC-FID)
Application in the environmental field 37PCB analysis with GC-ECD detection
Summary 39
Chromatography Volume 2I. OVERVIEW
8
as chromatography is a well-established technique, employedin many applicational fields suchas environmental, food, petro-chemical and pharmaceuticalanalysis [1].
The capillary columns used mayhave solid fillings (PLOTcolumns) or a liquid film as a stationary phase. The latter areemployed in gas liquid chroma -tography (GLC), which is thesubject of the present booklet. In GLC, analyte separation isachieved by means of a dynamicthermal equilibrium between thestationary phase (liquid) andmobile phase (gas). The dimen-sions of the columns employed,which vary depending on theapplication, generally range frominner diameters between 0.25 to0.53 mm and lengths from 25 to60 m. The film thickness select-ed, dependent on the boilingpoints of the analytes, usuallyranges from 0.25 to 1 μm (insome cases up to 3 μm or evenmore).
Nowadays in daily routine work,apart from increased analyticalsensitivity, demands are alsomade on the efficiency in termsof speed of the laboratory equip-ment. Regarding the rapidity ofanalysis, two aspects need to beconsidered:
1. The costs in terms of timerequired, for example in qualitycontrol analysis
2. The efficiency of the analyticalequipment employed
It is well known that analyticaltime costs may be very high instandard GC applications. In thelast few years, fast GC methodshave been shown to be suitablefor routine laboratory work. Inthis respect, the employment ofnarrow-bore columns is of par-ticular interest [2 -5].
These analytical tools enable aconsiderable reduction of analy-ses times while peak resolution isessentially maintained. Anotherapproach worthy of mention isthe use of multicapillary columns[6-8] with more than 900 capil-lary tubes with inner diametersof about 40 μm.
The present overview is focussedon the use of narrow-borecolumns in fast gas chromatogra-phy. As mentioned, this type ofapproach enables substantialreductions in GC run-timeswithout compromising the chro-matographic resolution.
An example of the validity ofthis approach is illustrated inFigure I.1 (lower chromatogram)
which reports a fast GC analysis(RTX-5 10 m, 0.1 mm ID, 0.4 μmfilm thickness) on a capillary col-umn test mixture (Grob test mix-ture, Macherey und Nagel). Theconditions were as follows: injec-tion mode: split (200 :1), averagelinear carrier gas velocity: 120cm/s (502 kPa, H2), temperatureprogram: from 140 °C at 60°C/min to 280 °C. A convention-al application (DB 5 30 m, 0.25mm ID, 0.25 μm film thickness)on the same matrix is also shownin the diagram (upper chro-matogram). As apparent, thedegree of separation is even bet-ter in the fast GC applicationwhile a speed gain of approxi-mately 8 times is attained.
However, the GC hardware hasto meet some requirements inorder to enable the efficient useof narrow-bore columns. In thefollowing chapters, general andfundamental issues relative tofast gas chromatography are dis-cussed (injection systems, capil-lary columns, carrier gas anddetector characteristics).
All the chromatograms shownwere recorded with a ShimadzuGC-2010 or GCMS-QP2010with an AOC-20i autoinjectorand GCsolution or GCMSsolu-tion software for data elabora-tion.
G
9
Chromatography Volume 2 I. OVERVIEW
6.0 7.0 8.0 9.0
Minutes
3
5
1
10.0 11.0 12.0 13.0 14.0 15.0
x 106
12
34
5
6 7
8
9
0.25 0.50 0.75 1.00
Minutes
2
4
0
1.25 1.50 1.75
x 104
FWHM = 0.48 s
12
3
4 5
6
78
9
FWHM = 3 s
6
Figure I.1:
Top: Chromatogram recorded with aGrob test mixture using a DB5 30 m,0.25 mm, 0.25 μm column.Bottom: Chromatogram recorded with the Grob test mixture using aRTX-5 10 m, 0.1 mm, 0.4 μm column.
Peak identification:1) n-Decane2) n-Octanole3) n-Undecane4) 2.6-Dimethylphenole5) 2.6-Dimethylaniline6) Decane acid methyl ester 7) Undecane acid methyl ester8) Dicyclohexylamine9) Dodecane acid methyl ester
N = 16 • ––––wb
tR2
Both the polarity and the boilingpoints of the sample analytesmust be considered during theselection of a specific stationaryphase material. Nowadays, sever-al phase materials are availablefor typical fast GC applications.
An important issue in fast GCregards the narrow-bore column
Chromatography Volume 2
1. Column and stationary phase
II. BASICS OF SEPARATION IN FAST GC
12
sample capacity, which is consid-erably lower than with conven-tional capillaries (this aspect willbe discussed later).
Apart from the choice of the sta-tionary phase, other importantcolumn parameters to be optimi -zed are length, internal diameterand film thickness.
The number of theoretical pla -tes N of a column is a measure ofits separation efficiency and isdefined as [9]:
2. Length and peak resolution
R = ––––4
� N �––––� – 1
k’2––––k’2 + 1
k’ = –––– ,tM
t’R � = –––– = ––––k’1k’2
t’R1
t’R2
L = column lengthN = –––––––HETP
L
R � � LR = ––––––––––––
_12 [wa + wb]
� t
where wb is the base peak widthand tR is the retention time. Peakresolution is of particular interestas the main aim is to reduce theanalysis time while maintainingthe column resolving power (Fig-ure II.2). The resolution betweentwo peaks is defined as:
Resolution (R) on the other handdepends on the capacity factor kand the plate number N in thefollowing equation [9]:
Wa Wb
�t
Figure II.1: Peak resolution parameters for two subsequent peaks in a chromatogram
with k 2 = capacity factor of themore retained solute and � = separation factor:
where tM is the time required foran unretained analyte to passthrough the column (generallyreferred to as dead time). Therelationship between the lengthof the column, N and height of atheoretical plate (HETP) is asfollows:
thus the resolution is propor-tional to the square root of thecolumn length:
Consequently, doubling the col-umn length achieves a gain ofresolution of a factor of only1.41. Therefore typical columnlengths in fast GC are within the10 - 15 m range.
• Column length for fast GC: 10 - 15 m
1
2
3
4
5
� � ––––dI4df
KD = k’ • �
13
Chromatography Volume 2 II. BASICS OF SEPARATION IN FAST GC
In general, the chromatographicpeak width increases with theretention time due to resistanceto mass transfer in the mobileand stationary phase contribu-tions. The influence of the innerdiameter with respect to the theoretical plate number is nowconsidered. The basic equationfor the retention of a substance is [9]:
The linear relationship betweenpeak resolution and stationaryphase thickness requires the lat-ter to be as thin as possible infast GC.
On the other hand, it should alsobe suited to the boiling points ofthe analytes (see also �). Sincethe column sample capacity alsodecreases with the film thickness(less amount of stationary phase),the split ratio needs to be adjust-ed depending on the concentra-tion of the sample. High splitratios ensure that the sample is
3. Inner diameter and peak resolution
4. Film thickness and peak resolution
HETPmin prop dI
where:dI,df = inner diameter/film
thickness of the column KD = distribution constant� = phase ratio
The distribution coefficient KDis dependent on the temperatureand the physical and chemicalproperties of the analytes andstationary phase. The retentiontime is therefore determined bythe inner diameter, film thicknessand length of the column forfixed material properties. Fortypical inner diameters and filmthicknesses such as 0.25 mm and
0.25 μm, the phase ratio is equalto approximately 250. This is atypical phase ratio value for amedium boiling point analyterange. In general, this parameteris used by the analyst to make afirst selection of the columndimension for a given distribu-tion of analyte boiling points.Column manufacturers usuallyoffer schematic tables in theircatalogues giving an overview oftypical � values for differentapplications. The value for � typ-ically varies between 50 (columnswith thick films for volatiles) and530 (wide-bore columns, highboilers).
For further elucidation, the socalled van Deemter curves [9]will be discussed in the followingchapter where HETP values fordifferent capillary columns areplotted versus the average linearcarrier gas velocity [u].
With regard to the experimentalcurves, the HETP values arederived from the peak widths(eq. 1 and 4) and are plotted as afunction of the average linearvelocity of the carrier gas. A typical curve is shown in Fig-
ure II.2 for 2 columns of differ-ing inner diameter [10]. The plateheight is characterized by a mini-mum value (HETPmin) at theoptimum linear velocity [9]. Both columns have a � value ofapproximately 250. For columnswith such a high phase ratio val-ue, the resistance to mass transferin the stationary phase can beneglected and HETPmin cantherefore be approximated to thecolumn ID [10]:
Average Linear Velocity (cm/sec)
Conventional and Ultra Fast
Van Deemter Curves C16 : 0 FAME
isothermal (150 °C)
0.00 50 100 150 200
0.10.20.30.40.50.60.70.80.91.0
HET
P (m
m)
30 m x 0.25 mm i.d. 0.25 μm film
10 m x 0.10 mm i.d. 0.10 μm film
Figure II.2: Van Deemter curves [10] for 0.25 mm and 0.1 mm ID columns. The phase ratio is 250 in both cases.
transferred rapidly from the glass insert to the column (seeIII. Injection).
As a conclusion, a typical filmthickness in fast GC of 0.1 μm(dI = 0.1 mm, � = 250) is used. In specific cases of highly vola -tile compounds film thicknessesof up to 0.4 μm (� = 62.5 for 0.1 mm dI) may be used.
• Columns in fast GC: film thicknesses are typically 0.1 μm (≤ 0.4 mm)
As seen from equation 7 and Fig-ure II.2 the minimum theoreticalplate height decreases linearlywith the capillary column innerdiameter. Furthermore, theascending part of the curves risesmore gradually, enabling theapplication of higher than opti-mum velocities with little loss interms of resolving power. Toconclude, the inner diameter ofsuitable columns for fast GC are:
• Columns for fast GC: inner diameter ≤ 0.15 mm
6
7
Chromatography Volume 2II. BASICS OF SEPARATION IN FAST GC
14
In fast GC applications, particu-lar attention must be paid to thetype of carrier gas used. The VanDeemter curves have differentminima for different gas types[9]. This factor is shown in detailin Figure II.3.
As can be seen, the minimumHETP for the hydrogen curve isattained at a higher average linearvelocity when compared to othercarrier gas types. In addition, theascending part of the curve risesmore gradually for the hydrogencurve. The data illustrated in Figure II.3 were derived from a0.25 mm ID column. In the caseof smaller inner diameters, asalready referred to, the ascendingpart rises even more gradually asmentioned in section 3.
5. Type of carrier gas and average linear velocity
The aforementioned issues have direct consequences on GC instru-mentation requirements such as injection system, carrier gas pneu-matics, column oven heating/cooling and detection. These aspects aresummarized as follows:
GC instrument requirements
• fast injection �� autoinjector, rapid injection, highsplit ratios (see chapter III.)
• reduced internal diameter �� large head pressure, high maxi-mum operational head pressure(up to 970 kPa)
• temperature program �� average linear carrier gas velocityshould be constant over theentire analysis (the constant lin-ear velocity mode requires apressure program)
• high separation efficiency �� linear temperature ramps arehigher compared to standard GC[70 °C/min linear and higher(ballistic)]
6. GC hardware requirements
In most applications, a lineartemperature program is employ -ed rather than isothermal analy-sis. Under isobaric conditions,
4020 60
N2
HETP (mm)
[u] cm/s
HETP – [u] curve for C17 at 175 °C with k = 4.95
Capillary column: OV-101 25 m, 0.25 mm ID. 0.4 μm film
0.4
0.8
80
He
H2
Figure II.3: Van Deemter curves measured using different carrier gases (Shimadzu Corporation, Kyoto, Japan)
average linear velocity decreaseswith increasing temperatures [9].This is due to the fact that thegas viscosity increases linearly
• high final temperature �� fast cooling of the GC oven(2 min from 280 °C to 100 °C)
• high concentration range �� high split ratios should be available (total flow of up to1200 mL/min)
• sharp peaks in fast GC �� high sampling rates and low filter time constants (250 Hz, 4 ms-conventional detectors)
The data in brackets refer to specifications of the GC-2010.
In the example illustrated in Figure I.1 (see page 9), a 120 cm/s aver-age linear velocity was applied. In order to obtain this velocity an initial pressure of 502 kPa was required at the initial temperature of140 °C. The gas chromatograph maintains the average linear velocityconstant over the temperature program (60 °C/min up to 280 °C).
This is achieved by the application of a pressure program automa -tically calculated by the GC software, and leads to a final pressure of 660 kPa at 280 °C. The cooling time for this analysis is below 2 minutes and the cycle time is under 4 minutes.
with temperature. Hence, theseparation efficiency is reducedas the temperature increases (theoptimum velocity is generallyselected for the initial tempera-ture) and the analysis time is alsoprolonged when compared toconstant velocity conditions.
Consequently, the gas chromato-graph should have an automaticpressure regulator in order tokeep the average linear velocityconstant over the entire tempera-ture program. This issue leadsdirectly to fast GC hardwarerequirements.
Chromatography Volume 2III. INJECTION REQUIREMENTS IN FAST GC
16
As can be observed in Figure I.1(see page 9), the peak widths athalf height (FWHM) are below0.5 s, much less if compared toconventional GC (about 3 s).The peak width is affected by thetime needed to transfer the sam-ple, vaporized in the glass insert,onto the column. The sampletransfer must be fast enough(movement of syringe plunger,purge speed of the glass insert).
In the case of a relatively slowsample transfer rate (such as insplitless injection, see FigureIII.1), this results in an initialsample band broadening on thefirst part of the column [11], which has a considerable effecton the whole separation efficien-cy. The well-known solventeffect indicated by Grob [11] isnot applicable as the column in -itial temperature has to be setbelow the solvent boiling point,making the analysis time longer,and in addition the film thicknessis very small. This requires thatthe glass insert is not overloadedwith vapor after liquid injection[11].
The sharpness of the peaks isalso affected. For example, a 1 μL n-hexane injection into a
Injection requirements in fast GCsplit/splitless injector results in avapor volume of about 175 μL at280 °C and a head pressure of100 kPa. In the case of acetone as solvent the vapor volumeunder the same conditions isabout 310 μL. Typical glass insertvolumes on the other hand arewithin 1-2 mL.
Whenever splitless injection isapplied in fast GC, a reducedvolume glass insert and highpressure injection are required.In general, sample transfer has aprofound influence on peakshape and needs to be optimized:
Samples with a high concentra-tion: high split ratio
At low concentration ranges (lowsplit ratio or splitless injection):
1. Use of glass insert with smallinner diameter to enhancepurging with carrier gas.
2. In the case of splitless injec-tion: high pressure injection.
1. Glass insert
Figure III.2 shows the effect of glass insert dimensions onpeak width. Here the Grob testmix was analyzed with a stan-
0.5 1.0 1.5
n-D
ecan
e
Glass insert with smallinner diameter
Standard glass insert
2.0 2.5 3.0 3.5 4.0
Dod
ecan
e ac
id m
ethy
l est
er
Dic
yclo
hexy
lam
ine
Und
ecan
e ac
id m
ethy
l est
er
Dec
ane
acid
met
hyl e
ster
Dim
ethy
lani
line
Dim
ethy
lphe
nole
n-U
ndec
ane
n-O
ctan
ole
Minutes
Figure III.2: Chromatograms relative to a Grob test mixture.Top: reduced ID (inner diameter) glass insert = 1 mm.Bottom: standard glass insert, 4 mm ID.
Figure III.1: Schematic of asplit/splitless injector (SPL-17): Fc: carrier gas, Fs: split flow, Fl: Flow through glass insert
Fc
Fs
Splitting point
Fl
Glass insert
Carrier Purge
Split
Column flow
17
Chromatography Volume 2 III. INJECTION REQUIREMENTS IN FAST GC
dard glass insert (4 mm innerdiameter – bottom) and a glassinsert with a 1 mm ID (top). Allother parameters remainedunchanged. In particular, thepeak widths of the less retainedvolatile components were nar-rower when the glass insertdimension was reduced. Theeffect of purging, which wasmore effective for the reduced IDglass insert, is clearly demon-strated. The split ratio was 40 :1in both cases. For increased splitratios the peak widths were alsonarrower when using the stan-dard glass insert.
In the chromatogram illustratedin Figure I.1 (see page 9), a 200 :1split ratio was selected. For peakno 1, the peak width measuredwith the standard glass insert wasFWHM = 0.42 s (split ratio200 :1) in contrast to 1.56 s inFigure III.2 – bottom (split ratio40 :1).
2. High pressure injection
As discussed in chapter II (sec-tion 5), the carrier gas averagelinear velocity has a great influ-ence on the column separationefficiency. The applied head pres-sure is therefore of great impor-tance after sample transfer to thecolumn. In this respect, it is pos-sible to apply an increased pres-sure just for sample transfer (typ-ically for 30 s to 1 min). Afterthis period the pressure isdropped automatically to theanalytical pressure to ensure peakresolution. This operation mode
is defined as high pressure injec-tion. Obviously, the purging ofthe glass insert is increased whencompared to standard conditions.In chapter V (section 1), whereanalysis of organophosphorouspesticides is discussed, high pres-sure injection (600 kPa) wasapplied in combination with asplitless injection technique inorder to optimize peak shapes.
3. Sample capacity of capillarycolumns
The maximum amount of samplewhich may be introduced onto acapillary column without over-loading is lower for narrow-borecolumns. When excessive sampleamounts are injected onto a cap-illary, peak fronting effects areobserved [9]. The generation ofasymmetric peaks may obviouslylead to a substantial loss in reso-lution. Consequently, the inject-ed sample amount must be belowa threshold value, as may beobserved in Figure III.3 [10].Here the plate number (calculat-ed from the peak width accord-ing to eq. 1) and hence separationefficiency is plotted as a functionof sample amount for twocolumns of different internaldiameter. It can be seen that forstandard columns (0.25 mm ID)up to 50 ng solute quantities maybe injected with little or no lossof separation efficiency. For anarrow-bore column with a 0.1mm ID, analyte amounts of up to1 ng may be injected without los-ing separation efficiency.
Amount injected (ng)
0.1 1.0 10.0
60
70
80
90
100
Plate Number X 100025 m x 0.25 mm i.d. 0.25 μm film thickness
10 m x 0.10 mm i.d. 0.10 μm film thickness
100.0 1,000.0
Figure III.3: Number of theoretical plates as a function of sample amount for two different columns
Chromatography Volume 2IV. DETECTOR REQUIREMENTS IN FAST GC
20
Filter time constant and sampling frequency
1. Conventional detectors
Signal height = 63.2 %
Time
Signal
Figure IV.1: Definition of filter time constant �
0.030 0.031 0.032
uV (x 10,000)
Minutes
0.0
0.5
1.0
1.5
2.0
0.033 0.034 0.035 0.036
4 ms
20 ms
50 ms
100 ms 200 ms
Filter time constant
Figure IV.2: FID signal of chlorodecane. The filter time constant � was changedfrom 4 to 200 ms. The sampling frequency was 250 Hz in all cases.
In this section, conventionaldetectors such as the flame ion-ization detector (FID), electroncapture detector (ECD), flamephotometric detector (FPD) andflame thermionic detector (FTDor NPD) are discussed.
Mass spectrometric detection isdescribed in the next section. As mentioned, typical fast GCpeak widths are 0.5 s or less. The detection system must meeta series of requirements in orderto maintain the peak width atcolumn outlet. In this section therequirements for FID are dis-cussed, but the principles arevalid for all of the conventionalGC detectors. In general, detec-tor peak broadening may becaused by two factors. Firstly,the possible presence of deadvolumes in the sample path.Modern GC detectors are gener-ally characterized by low deadvolumes. Furthermore, make-upgas is employed to optimize peakshapes through an additionalflow (typically 30 mL/min). Formass flow dependent detectors
(FID, FPD, FTD) this featureenhances sensitivity [9].
In modern GC, the detector sig-nal is digitalized in the detectorelectronics and the A/D convert-er passes digital data to the PC.Due to the presence of RCchain-containing circuits, anysignal amplifier has a certainresponse time with the presenceof some smoothing effect. Thisshould not influence the peakwidth which is determined bythe chromatographic process prior to the detector (injectionsystem, column). The parameterof detector electronics regardingthe response time is usuallyreferred to as filter time constant(FTC) [12]. The electronic detec-tor signal amplifier cannot followa signal in an infinitely fast way.It needs a certain time to followrapid signal changes. In FigureIV.1 this parameter is visualizedgraphically [12]. A theoreticalstep increase of signal is shownhere and the response of detectoramplifiers indicates that it fol-lows the signal with some time
delay (1-1/et). The time intervalwhich the detector electronicsneeds to reach the (1-1/e) frac-tion of such a step signal isdefined as filter time constant.
This parameter should beselectable [12] together with thenumber of data points recordedacross a peak. In an experiment,the influence of this parameteron real peak shapes was meas-ured. The effects of the injectionsystem and column on bandbroadening were reduced as fol-lows:
1. A 0.1 mm ID glass insert and a0.2 μL injection volume of justone compound (chlorodecane)with a high split ratio (600 :1).
2. Use of a 1 m column with a 0.1 mm ID and 0.1 μm film thickness; isothermal analysis at 200 °C.
3. Detector make-up gas flow of 60 mL/min.
Chlorodecane was diluted inmethanol. Figure IV.2 shows thechlorodecane peak which elutesat 0.0332 min under isothermalconditions. The filter time con-stant � was varied from 4 ms to200 ms. It can be clearly obser -
Filter time �
21
Chromatography Volume 2 IV. DETECTOR REQUIREMENTS IN FAST GC
2. Mass spectrometric detector (quadrupole MS)
0.2 0.4 0.6 0.8 1.0 1.2 1.4
Filter time constant �
� = 100 ms
1.6 1.8 2.0 2.2 2.4 2.6 0.2 0.4 0.8 1.0 1.2 1.4
� = 10 ms
1.6 1.8 2.0 2.2 2.4 2.6
Figure IV.3: Chromatograms relative to a kerosene sample recorded at a filter time constant � of 100 ms and 10 ms respectively
5 ms 10 ms
Nor
mal
ized
SN
rat
io
Time constant
0
0.2
0.4
0.6
0.8
1
1.2
20 ms 50 ms 100 ms 200 ms 500 ms 1 s 2 s4 ms
Half width 10 s
Half width 1 s
Half width 0.1 s
Figure IV.4: Signal-to-noise ratio as a function of the filter time constant and for different peak widths (source: GC-2010 operation manual)
ved that increases in the filtertime constant affects both thepeak retention time and shape aspredicted from theory [12]. Thefull width at half maximum for a� value of 200 ms (a standardparameter in many GC systems)is about 3 s, whereas for a � valueof 4 ms the FWHM value isabout 40 ms. The effect of the filter time constant on peakshape obviously has a drasticeffect on resolution as shown inFigure IV.3. The figure illustratestwo kerosene very fast GC chro-matograms with applied filtertime constants of 100 and 10 msrespectively. The improvement interms of peak resolution is evi-dent in the 10 ms FTC analysis.This experiment shows theimportance of detector parameteroptimization in terms of peakresolution. In addition, the FTChas a considerable influence onthe signal-to-noise ratio (S/N).Figure IV.4 shows the normal-ized signal-to-noise ratio as afunction of the filter time con-stant. These curves are plottedfor different peak widths. For apeak width of 0.1 s the optimumfilter time constant is 20 ms withregard to the S/N ratio. Theoptimum FTC value under suchconditions is also dependent onthe sample concentration andranges between 4 and 20 ms.
A further electronic parameter ofimportance is the sampling rateor, in other words, the number ofdata points acquired per secondwhich are transferred from theA/D converter to the PC. Thenumber of points must be suffi-cient for the proper reconstruc-tion of a Gaussian peak (mini-mum 10 to 15 data points). If thenumber of data points is too low,data reproducibility is reduced.The filter time constant and thesampling frequency parametersnecessary for the range of peakwidths observed in gas chro-matography were summarized byJ.V. Hinshaw [12]. The latterreported that for peak widthsbetween 0.1 and 0.5 s the filtertime constant and sampling rateshould be about 10 ms and 20 to100 Hz respectively.
Minutes Minutes
The most common mass spectro-metric detector used in gas chro-matography is the quadrupoleMS. With regard to fast GC/MSanalysis, the MS detector obvi-ously needs to be fast enough forthe rapid eluting analyte bands.For identification purposes, GC/MS applications are carried
out in the full scan mode inorder to perform a MS librarysearch. In this operation mode,every data point corresponds to acomplete mass spectrum over themass range selected (scan).
After peak assignment, the SIMmode (selected ion monitoring)
may be used in order to increasethe analytical sensitivity.
In order to attain this data, twoparameters of the GC/MS hard-ware must be considered: firstly,the mass range selected by theanalyst must cover all the analyteion fragments in order to �
Chromatography Volume 2IV. DETECTOR REQUIREMENTS IN FAST GC
22
3.87 3.88 3.89 3.90 3.91 3.92 3.93 3.94 3.95 3.96 3.97 3.98
6,086,2713.922
TIC*1.00
Minutes
m/zR.Time: 3.921 (Scan#: 3026)
m/zR.Time: 3.922 (Scan#: 3027)
Figure IV.5: Definition of scan speed, inter-scan delay and sampling frequency for a quadrupole MS
100
40 80 120 160 200 240 280
100
40 80 120 160 200 240 280
Scan N
m1
Intervall I:
1/I = sampling frequency
Inter-scan time
RF setup
m2 Scan N +1
2.4min. 2.5 2.4min. 2.5
Linalool
61,307,110 61,307,110TIC TIC
2.4min. 2.5
61,307,110 TIC
20 60 100 140 180
m/z
WileySI = 97
20 60 100 140 180
m/z
WileySI = 98
20 60 100 140 180
m/z
WileySI = 97
Figure IV.6: Linalool peak generated bya narrow-bore column (FWHM 0.6 s)and the spectra recorded at the begin-ning, apex and end of the peak afterbackground subtraction. Spectra simi-larities are 94, 96, 94 using the WileyMS library.
perform a library search for com-ponent identification (scan). Inthis case, the scan speed shouldbe high enough so that the rela-tive intensities of the fragmentsdo not change significantly frompoint to point across a narrowpeak. In this respect, a low scanspeed would result in low qualitylibrary matches. Secondly, thesampling frequency, which cor -responds to the number of scans(or the number of SIM datapoints) per second, is, as afore-mentioned, a fundamental para -meter. Fast GC/MS applicationsrequire a high quadrupole scanrate (up to 10,000 amu/s) and, inaddition, a low inter-scan deadtime (reset time of the electro -nics) so that the sampling rate may be sufficiently high (up to50 Hz/100Hz in scan/SIM modesfor the GCMS-QP2010). Thissituation is shown in Figure IV.5.
The true scanning speed of thequadrupole is calculated by thescan range divided by the intervaltime minus the RF setup timewhich ranges from 5 to 10 ms for the GCMS-QP2010. In Fig-ure IV.6 this is demonstrated byexperimental data. A linaloolpeak was scanned with a 20 Hzscanning frequency and a select-ed mass range of 30 to 350 amu.The analysis was carried out witha SPB5 10 m, 0.1 mm, 0.1 μm col-umn. The peak width at halfheight is about 0.6 s and about1.2 s at the base resulting in 24 data points across the peak.As can be seen, the quality of thespectra in three points across thechromatographic peak is good (a Wiley library search was per-formed). The spectra similarities,94, 96, 94, indicate that there isno skewing.
Chromatography Volume 2V. APPLICATIONS IN FOOD ANALYSIS
24
1. Organophosphorous pesticides in food matrices
Organophosphorous pesti-cides (OPPs) are widelyused in agriculture
in order to protect a broad rangeof products. Unfortunately, thepresence of these toxic compoundsis often detected in foods.
The maximum allowable concen-trations of OPPs in foods is,
3.00
5
Minutes
Trichlorphon
1.75
1.50
1.25
1.00
0.75
0.50
0.25
0.00
2.00
10 15 20
RTX-5: 10 m, 0.1 mm, 0.4 μm
RTX-35: 30 m, 0.25 mm, 0.5 μm
Phosphamidon (E)
Azinphos - MeEthionFenthionPhosphamidon (Z)
Carbophenothion
ProfenophosMethidathion
Tolclophos - Me
DisulfotonDioxathion
3.00
1.50
1.25
1.00
0.75
0.50
0.25
0.00
1
4
Minutes
6 8
1.75
2.00
2.25
2.50
2.75
11
10
9
8
7
6
5
4
3
2
13
uV (x 1,000.000)
Figure V.1: Top: conventional GC chromatogram of organophosphorous pesticide standards in a tomato extract. Concentrations vary beween 125 and 1,000 pg. Bottom: fast GC organophosphorous pesticide analysis carried out with aRTX-5 : 10 m, 0.1 mm, 0.4 μm column.
Peak identification: 1. Dichlorphos, 2. Trichlorphon, 3. Dioxathion, 4. Phosphamidon (E), 5. Disulfoton, 6. Phosphamidon (Z), 7. Tolclophos methyl, 8. Fenthion, 9. Methidathion, 10. Profenophos, 11. Ethion, 12. Carbophenothion, 13. Azinophos methyl
however, strictly controlled [13].In Germany, the well-knownDFG S19 (Deutsche Forschungs-gemeinschaft) is employed forthe determination of traces ofthese substances in foods. ThisStandard is also present in Euro-pean regulations as DIN EN1528-3, DIN EN 12393-2. Inaddition, these methods define
the procedure for the determina-tion of organochlorine, nitrogen-containing and other types of pes-ticides. The procedure steps aremainly: 1. extraction and partition,2. gel permeation chromatography,3. mini silicagel chromatographyand 4. analysis with GC-FTD,GC-FPD or GC/MS.
A chromatogram relative to anOPP-spiked (amount rangebetween 125 and 1,000 pg) toma-to matrix is shown in Figure V.1.The column used was a RTX-3530 m, 0.25 mm, 0.5 μm. The fig-ure also shows the resultobtained with a RTX-5 (5 %phenyl), 10 m, 0.1 mm, 0.4 μmcolumn under fast GC condi-tions [14] (temperature programfrom 100 °C (1 min) at 80 °C/min to 220 °C at 10 °C/min to240 °C at 50 °C/min to 320 °C;carrier gas: H2 (120 cm/s). Theinjection volume was 2 μL in thesplitless mode. In order toincrease the sample transferspeed (see above) a 600 kPa highpressure injection was appliedfor 1 minute. The peak widths(FWHM) are approximately 3 sin the standard analysis and,below, 0.5 s in the chromatogramobtained with the narrow-borecolumn. As the area in general isdetermined by the sampleamount which was not changed(both runs were carried out inthe splitless mode) the signal-to-noise ratio was higher in the fastGC experiment. The retentiontime of azinophos methyl was 22minutes in the standard analysisand 9 minutes in the fast result.
A chromatogram derived from ablack tea analysis is shown inFigure V.2. The analytical resultindicated a contamination ofdioxathion (225.2 pg), phos-phamidon (67.1 pg), disulfoton(217 pg), dichlorphos (243 pg),ethion (145.1 pg) and azinophosmethyl (244.8 pg).
The limit of detection in a fastGC application may be derived,for example, from the results
12
MethamidophosMethomylDiazinonChlorpyriphos-OEtCaptanEndosulfan sulfatePiperonyl butoxidePotasan
--
169313, 315, 212, 214
150, 14997, 386
-328, 192
NCI dominant Peaks(Da)Substance
25
Chromatography Volume 2 V. APPLICATIONS IN FOOD ANALYSIS
7.0 8.0 9.0
Dic
hlor
phos
uV (x 1,000,000)
Minutes
0.00
0.25
0.75
0.50
1.25
1.00
1.50
6.05.04.03.02.0
Azi
noph
os m
ethy
l
Ethi
on
Phos
pham
idon
(Z)
Dis
ulfo
ton
Phos
pham
idon
(E)
Dio
xath
ion
Figure V.2: Fast GC chromatogram of a black tea sample. For concentrations see text.
0 1 2 3 4 5
Ethy
lpar
athi
on
Minutes
95,00090,00085,00080,00075,00070,00065,00060,00055,00050,00045,00040,00035,00030,00025,00020,00015,00010,0005,000
0- 5,000
Bro
mop
hos
met
hyl (
i.s.)
Figure V.3: Fast GC chromatogram of an olive oil sample. For the quantitativedetermination of ethylparathion, bromophos methyl was added as internal stan-dard; concentration ethylparathion: 3 ppb.
4.0 5.0 6.0
Dia
zino
n
(x 10,000,000)
0.1
0.3
0.4
0.5
0.6
0.7
0.8
1.0
0.9
1.2
1.1
1.3
7.0
Pota
san
Pipe
rony
l but
oxid
e
Endo
sulfa
n su
lfate
Capt
an
Chlo
rpyr
opho
s
100 °C init. temp.
He 50 cm/sec
NCI 50 pg Splitless
EI with combi source 5 ng Split 20 : 1
100.0 125.0 150.0
0.0
2.5
5.0
7.5
175.0 200.0 225.0 250.0 275.0 300.0 325.0 350.0 375.0 400.0
EI
(x 10,000)
m/z 407.00 Abs. Inten
Base Peak: 272 / 79,712
102 121
0.2
133 143 157169
170 193 206217 226
231
229
237
242 263
270
272
276279 289
301 311 341349357 379
386
387
390395
0 Rel. Inten 0.00
100.0 125.0 150.00.0
2.5
5.0
175.0 200.0 225.0 250.0 275.0 300.0 325.0 350.0 375.0 400.0
(x 1,000,000)
m/z 351.70 Abs. Inten
Base Peak: 97 / 2,332,743
145
97
337392388371
11599 185158
386
209200 219 234 290270254 306 314350
86.363 Rel. Inten 3.70
Scan 50 - 400 amu
2000 amu / sec
Figure V.4: Total ion chromatograms (TIC) of an OPP standard mixture analysis carried in the EI and NCI modes
shown in Figure V.3 [15]. Thedata shown were obtained froman olive oil sample. A 3 ppb con-tamination of ethylparathion wasdetermined by using internalstandard (bromophos methyl)calibration. During samplepreparation, 2 g of olive oil wereadded to 2 mL of hexane. Theinternal standard (0.01 ppm bromophos) was then added. Aliquid extraction with acetonitrilewas then carried out and, follow-ing this, 7.5 mL was taken fromthe acetonitrile phase and dried.
Finally, 2 mL of hexane wasadded before GC injection.
GC/MS data:
In GC/MS, the electron impact(EI) mode for analyte ionizationis most widely employed. How-ever, in recent years the chemicalionization mode has becomemore popular. Sensitivity may beconsiderably enhanced for elec-trophilic molecules by negativechemical ionization (NCI) �
94, 141, 126, 111, 110, 12858, 105, 88, 73, 115, 162
304, 179, 137, 152, 276, 199314, 187, 258, 244, 97, 125, 35179, 117, 149, 264, 301, 182, 236
229, 272, 239, 207, 170, 387176, 177, 149, 178
192, 328, 176, 300, 272, 148
EI Peaks (Da)
Table V.1: EI and NCI dominant peaks observed with the organophosphorouspesticides
NCI
Equality
AldinThiobencarbPermethrinPyributicarbp,p'-DDDChlorpyrofosBitertanolIsofenphosQuinalphosp,p'-DDEPirimiphos methylp,p'-DDTHalfenproxEthoprophosDichloros
p,p'-DDTDimethoate�-LindaneCafenstroleFluvalinate-1Pyrifenox-EDiflufenicanHexaconazoleThiometonDiazinonProthiofosDifenoconazolePyridabenTeraconazoleChinomethionateBifenthrinInabenfideMyclobutanil
BifenoxFenvalerateChlorfenapyrCyfluthrinPyrethrin-2MEPEPNFenpropathrinAcrinthrinTrifluralinPhosaloneCyhalothrin�-CVPFlucythrinateButamifos�-BHCCaptanImibenconazole
> 10 times
MepronilPropiconazolLenacilTebucanzoleEtoxazoleTebufenpyradPyriproxyfenPyrimidifenMefenacetTriadimenolEtrimfosChlorobenzilateEsprocarbThenylchlorTricyclazoleIsoprocarb
Chromatography Volume 2V. APPLICATIONS IN FOOD ANALYSIS
26
3.0 4.0 5.0
Chlo
rpyr
ofos
(x 1,000,000)
0.00
0.50
0.75
1.00
1.25
1.50
1.75
2.25
2.00
2.75
2.50
0.25
3.00
3.25
6.0 7.0 8.0 9.0
Endo
sulfa
n su
lfate
Black Tea
Mix 0.5 pg
Figure V.5: Black tea and standard compound mixture (0.5 ppb of each compound) TIC chromatograms (NCI mode)
> 10 times
DichlofluanidFenalimolPendimethalinDieldrinEdifenphosTefluthrinCypermethrinThifluzamideAcetamipridPAP, PhenthoatePyrethrin-1EndrinDeltamethrin�-BHC�-BHC
Dicofol BunkaibutuMalathionParathionDimethylvinphosPyrifenox-ZParathion methylKresoxim methylFolpetCyanazineCadusafosTolclofos methylPretilachlorMalathionCaptafolFlutolanil
FosthiazateAcephateTerbacilPyraclofosPenconazolUniconazolepFensulfothionCyhalofop buthylButachlor
ChlorfentezinedegMetamidophosEPTCPropamocarbChlorprophamTerbufosBenfuresateDimethenamidAlachlorMetolachlorDiethofencarbFenthionFludioxonilMethoprenePaclobutrazolFlusilazoleCyproconazole
2 - 10 times 2 - 10 times
Comparison of the sensitivities in EI and NCIHigher sensitivity in NCI mode Higher sensitivity in EI mode
Table V.2: Classification of sensitivities with EI and NCI modes for pesticides
using iso-butane, methane orammonia as reagent gas. Theselectivity is also helpful whenpeak coelutions are observedwith other matrix components[16]. In order to identify pesti-cides the EI mode of theNCI/PCI/EI combi source wasused in these experiments. TheNCI masses of the compoundswere then determined by com-parison of the EI and NCI scandata. In Figure V.4, the GC/MSresults attained for a standardusing a RTX-5 10 m, 0.1 mm, 0.1 μm column and using bothmodes are shown. The resultingspectra similarity is comparablewith the result obtained usingthe standard method. In thisexperiment, the scan range was50 to 400 amu with a scan fre-quency of 20 Hz. The amountsof the two standards were 5 ng(EI) and 50 pg (NCI). While theMS response is enhanced drasti-cally for diazinon, chlorpyrofos,endosulfan sulfate and potasan,the signal increase is less for cap-tan, while for piperonyl butoxidethere is almost no signal at all. To illustrate the different spectragenerated by the two modes, thecorresponding spectra of endo-sulfan sulfate are also shown.
With regard to EI, a typical frag-mentation is observed; in the caseof NCI only the masses 384, 386,388 together with a dominant 97fragment are present. Table V.1reports data concerning differentdominant fragments observed inEI compared with NCI for thecompounds measured here. Fig-ure V.5 shows a TIC chromato -gram of a real sample (black tea)indicating a contamination ofchlorpyrofos and endosulfan sulfate. Also shown is a chro-matogram relative to a standardmixture containing 0.5 pg of eachcomponent.
Table V.2 shows a list of pesticidesand a classification in terms of sen-sitivity in each mode [16]. As NCIis characterized by a very highselectivity, it can also prove use-ful when matrix componentsinterfere with the peaks of inter-est. For a more complete over -view of pesticides screened witha combination of EI and NCI,please consult reference [16].
In conclusion, a combination ofEI and NCI as well as mass spec-tra library search for both modesis a powerful tool for pesticideanalysis.
27
Chromatography Volume 2 V. APPLICATIONS IN FOOD ANALYSIS
2. The analysis of organochlorine pesticides in food matrices using GC-ECD and NCI GC/MS
Organochlorine (OCP)pesticides are commonlymeasured using ECD or
MS (NCI mode) detectors.
In conventional GC analysisusing standard columns of about30 m length with a 0.25 mm innerdiameter and 0.25 μm film, thetypical run time for an OCP stan-dard solution containing 23 com-pounds is about 30 minutes (Fig-ure V.6; for quantitative resultsrefer to Table V.3). The retentiontime of p,p’-DDD is about 21.5minutes. The column used in thisapplication was a 5 % phenyl; thetemperature program was as fol-lows: from 100 °C (1 min) to 170°C, at 50 °C/min (1 min), then to220 °C at 5 °C/min, then to 260°C at 10 °C/min, then to 280 °C(10 min) at 20 °C/min using N2 ascarrier gas with an initial pressureof 77 kPa corresponding to anaverage linear velocity of 23 cm/s.The injection was carried out inthe splitless mode (1 μL).
The application was then carriedout using a fast GC method witha CPsil 8 9 m, 0.1 mm, 0.1 μmcolumn and H2 as carrier gas. The result is shown in Figure V.7.As can be seen, all 23 compoundsare better separated and the reten-tion time of p,p’-DDD was lessthan 3.6 minutes. The tempera-ture program applied was as fol-lows: from 80 °C (1 min) to 280°C (3 min) at 60 °C/min with aninitial head pressure of 324 kPaand an average linear velocity of100 cm/s (H2 ) which was con-stant during the entire analysis.The selected filter time constantand sampling frequency were 20 ms and 63 Hz respectively. Theinjection volume was 1 μL with asplit ratio of 40 :1. The ECDmake-up gas flow was 80 mL/min.The signal-to-noise ratio of �-HCH in the rapid application isabout 440 :1 (split) while the samemeasured 220:1 in the splitlessconventional application, indica -ting an increased sensitivity.
10.0 12.5 15.0 17.5 20.0 22.5
Tecn
azen
e
Minutes25.0 27.5
- 0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0uV (x 1,000)
Mir
ex
p,p’
-DD
D
cis-
Chlo
rdan
etr
ans-
Chlo
rdan
eB
rom
opho
s et
hyl
tran
s-H
epta
chlo
roep
oxid
eci
s-H
epta
chlo
roep
oxid
eIs
odri
neB
rom
opho
s m
ethy
lIs
oben
zaneH
epta
chlo
r
Pent
achl
orbe
nzen
e
Ald
rine
Pent
achl
oran
iline
delta
-HCH
Lind
ane
beta
-HCH
HCB
Pent
achl
oran
isol
e
alph
a-H
CH
Ben
flura
line
Figure V.6: Conventional GC analysis of an OCP standard mixture (23 compounds) using a RTX-5 30 m, 0.25 mm ID, 0.25 μm column. For operational conditions refer to text.
2.25 2.50 2.75 3.00 3.25 3.50
Tecn
azen
e
Minutes3.75 4.00
- 0.250.000.250.500.75
1.251.501.75
2.252.50
3.003.25
3.754.00
4.504.755.00
5.50uV (x 10,000)
Mir
ex
p,p’
-DD
Dci
s-Ch
lord
ane
tran
s-Ch
lord
aneB
rom
opho
s et
hyl
tran
s-H
epta
chlo
roep
oxid
eci
s-H
epta
chlo
roep
oxid
eIs
odri
neB
rom
opho
s m
ethy
l
Isob
enza
ne
Hep
tach
lor
Pent
achl
orbe
nzen
e Ald
rine
Pent
achl
oran
iline
epsi
lon-
HCH
Lind
ane
beta
-HCH
HCB
Pent
achl
oran
isol
e
alph
a-H
CH Ben
flura
line
2.00
5.25
3.50
1.00
2.75
4.25
Figure V.7: Fast analysis of the OCP standard mixture containing 23 compounds with a CP SIL 8 9 m, 0.1 mm, 0.1 μm column.For retention times and concentrations see Table V.3.
The full width at half maximumof �-HCH, for example, is about 0.5 s which is a typical value forthese types of columns, demon-strating the suitability of theECD-2010 for fast analysis in the
field of organochlorine pesticides.The limit of detection for �-HCH(considering a signal-to-noise ratioof at least 3) was about 0.1 ppb.The method was then applied to a grape sample. The resulting
chromatogram is shown in FigureV.8. Contaminations of chlor -pyrofos at a concentration of 0.53 ng/mL (corresponding to 0.48 ng/kg grapes) and �
TargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTarget
RT2.232.372.58
2.5482.6132.6312.5982.6622.8182.6832.842
2.963.1093.0653.1393.1633.1963.2063.2913.3253.4693.5083.961
01 Pentachlorobenzene02 Tecnazene03 Benfluraline04 �-HCH05 HCB06 Pentachloroanisole07 �-HCH08 Lindane09 �-HCH10 �-HCH11 Pentachloroaniline12 Heptachlor13 Aldrine14 Isobenzane15 Bromophosmethyl16 Isodrine17 cis-Heptachloroepoxide18 trans-Heptachloroepoxide19 Bromophosethyl20 trans-Chlordane21 cis-Chlordane22 p,p’-DDD23 Mirex
Chromatography Volume 2V. APPLICATIONS IN FOOD ANALYSIS
28
1.5 2.0 2.5 3.0 3.5 4.0
Minutes
4.5- 0.5
0.0
1.0
1.5
2.0
3.0
3.5
4.5
5.0
6.0
6.5
7.5
8.0
8.5
9.5
uV (x 10,000)
Cypermethrin
Chlorpyrofos
2.5
9.0
5.5
0.5
4.0
7.0
Figure V.8: Fast GC-ECD result for a grape sample: chlorpyrofos 0.53 ng/mL (corresponds to 0.48 mg/kg grapes) and cypermethrin 0.55 ng/mL (corresponds to 0.5 mg/kg)
1.25 1.75 2.25 2.75 3.25 3.75
Tecn
azen
e
Minutes
4.25 4.50
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
uV (x 100,000)M
irex
cis-
Chlo
rdan
e
tran
s-Ch
lord
ane
Bro
mop
hose
thyl
tran
s-H
epta
chlo
roep
oxid
eci
s-H
epta
chlo
roep
oxid
eIs
odri
neB
rom
opho
smet
hyl
Isob
enza
ne
Pent
achl
orob
enze
ne
Ald
rine
Pent
achl
oroa
nilin
ede
lta-H
CHLi
ndan
ebe
ta-H
CH
HCB
Pent
achl
oran
isol
e
alph
a-H
CHB
enflu
ralin
e
4.003.503.002.502.001.50
p,p’
-DD
D
Hep
tach
lor
epsi
lon-
HCH
Lind
ane
Figure V.9: Fast GC chromatogram relative to an OCP standard mixture. Injection volume: 1 μL, in the splitless mode, high pressure pulse 400 kPa. Column: RTX-5 10 m, 0.18 mm, 0.4 μm.
cypermethrin at a concentrationof 0.55 ng/mL (corresponding to0.5 ng/kg) were found.
In order to apply a splitless injec-tion technique, high pressureinjection in combination with aslightly thicker film needs to beused (Figure V.9). For this appli-cation, a 10 m, 0.18 mm, 0.4 μm (5 % phenyl) column wasused with a 100 °C initial temper-ature for 1 min then at 60 °C/minup to 280 °C (3 min). The carriergas was H2 and the linear velocitywas 120 cm/s over the entire run.All other parameters wereunchanged. Again considering �-HCH, the calculated detectionlimit is about 0.01 ppb in thiscase.
The determination of organochlo-rinated pesticides in food matricescan be performed very well usingfast GC-ECD with the ECD-2010. The measured detectionlimits were below 0.1 ppb for sev-eral compounds using a split of40 :1 and about 0.01 ppb using thesplitless technique.
In the case of GC/MS, negativechemical ionization may be con-sidered as ideal in order to pro-mote a thermal electron captureprocess inside the ion sourceusing a reagent gas (methane).Low energy secondary electrons
originate from the impact of theelectrons emitted from the MS fil-ament onto the reagent gasmolecules. The ions producedthen pass through the mass filterjust as in EI. Only the polaritiesof the instrument are reversed. Itis very interesting to compare theresults obtained by employingNCI GC/MS with those obtainedwith the ECD. Figure V.10 illus-trates data relative to six organo -
Table V.3: Concentration of the OCP standard. Chromatogram shown in Figure V.6.
chlorine compounds recorded inthe SIM mode. The concentra-tions employed were between 0.4 and 0.6 ppb. The signal-to-noise ratios observed for t-hepta -chloroepoxide and p,p’-DDE are28:1 (mass 71) and 210:1 (mass70) respectively. If a minimumsignal-to-noise ratio of 3 :1 is con-sidered, the limits of detection aresimilar to the ECD results. Thecalibration curve for t-hepta -
chloroepoxide is also illustrated,showing a good correlation.
In conclusion, EI and NCI incombination is a very powerfultool for screening of organophos-phorous or organochlorine pesti-cides.
Conc (ppb)21.322.552.822.124.120.620.428.823.2
126
30.421.7
522.6
22.042525
50.3655
22.421.84
29
Chromatography Volume 2 V. APPLICATIONS IN FOOD ANALYSIS
4.9 5.15.0 5.2
25,424
316
318
237
ID#:1
Name: Oxychlordane
R.T.: 5.008
Conc.: 0.44200 ppb
5.8 6.05.9 6.16.1
45,640
404 406
242
ID#:5
Name: Endosulfan I
R.T.: 5.888
Conc.: 0.60456 ppb
4.9 5.0 5.1 5.2
30,396
71
280
282
ID#:2
Name: t-Heptachloroepoxide
R.T.: 5.062
Conc.: 0.48018 ppb
5.95.8 5.0 6.0 6.16.1
54,371
308
272
274
5.3 5.4 5.4
66,700
242
406
244
ID#:3
Name: Endosulfan II
R.T.: 5.350
Conc.: 0.51491 ppb
6.65.6 6.7 6.8 6.9 7.0
11,109
70
308
310 ID#:7
Name: p,p’-DDE
R.T.: 6.597
Conc.: 0.63673 ppb
5.5 5.6 5.7
7,135
346
237
344 ID#:4
Name: Dieldrin
R.T.: 5.585
Conc.: 0.45632 ppb
2.0 5.0 ppb0.0
[*10 ^4]
Calibration
ID#:2
Name: t-Heptachloroepoxide
Mass: 71.00
0.0
2.0
5.0
Figure V.10: NCI (methane) GCMS data recorded in SIM mode (figures on the graphs refer to the selected mass traces) with an organochlorine standard containingoxychlordane, t-heptachloroepoxide, endosulfan I, dieldrin, endosulfan II, endrin and p,p’-DDE. Concentration of each compound is 0.4 to 0.6 ppb.
ID#:6
Name: Endrin
R.T.: 6.011
Conc.: 0.50638 ppb
Chromatography Volume 2V. APPLICATIONS IN FOOD ANALYSIS
30
1.61.31.12.80.23.5
11.80.91.1
32.01.80.60.3
10.124.2
2.70.6
C4 : 0C6 : 0C8 : 0
C10 : 0C10 : 1C12 : 0C14 : 0C14 : 1C15 : 0C16 : 0C16 : 1C17 : 0C17 : 1C18 : 0C18 : 1C18 : 2C18 : 3
1.41.41.12.70.23.5
11.61.01.3
32.01.80.70.3
10.724.2
3.00.5
Conventional(run time: 33.0 minutes)
Ultra-fast(run time: 2.0 minutes)
3. Analysis of butter fatty acid methyl esters (GC-FID)
Fatty acids (FAs) have essen-tial physiological functionsin the human body and
therefore have an important influ-ence on human health.
In particular, �3 unsaturated fattyacids are of great importance asthey can not be produced by thehuman metabolism. FAs such aseicosanic acid (EPA) and docosa-hexanioc acid (DHA) have a ben-eficial influence on immunologi-cal, heart and allergic diseases,diabetes and cancer [17]. As aconsequence, the determination oflipidic profiles in foods is veryimportant. The composition offats and oils in natural matricesmay be very complex and conven-tional GC separations can there-fore be time-consuming. It mustbe added that prior to GC injec-tion, the FAs must be convertedinto their respective methyl estersin order to increase volatility anddecrease polarity.
A conventional GC applicationcarried out with butter FAMEs isillustrated in Figure V.11 [18]. The separation may be consideredto be quite complex and with thecolumn type used (Carbowax 30 m, 0.25 mm ID, 0.25 μm film),the total analysis time exceeds 30 minutes.
In contrast, the result obtainedwith a fast GC method is shownin Figure V.12 (Supelcowax 10 m,0.1 mm ID, 0.1 μm film) [18]. Theresolution may be considered tobe approximately the same butthe total analysis time is reducedby a factor of 17. The parametersapplied in the conventional analy-sis were as follows: injection vol-ume: 1 μL (1 :10 in hexane); splitratio: 1:50 (250 °C); temperatureprogram: 50 °C to 250 °C at 3.0°C/min; head pressure: 53 kPa atconstant average linear velocity;carrier gas: H2; [u]: 36.2 cm/s;detector: FID (250 °C), H2: 50mL/min, air: 400 mL/min, make-up: 50 mL/ min kPa (N2); sam-
5.0 15.0 25.0Minutes
0 e3
50 e3
100 e3
150 e3
200 e3
250 e3
300 e3
350 e3
uV
30.020.010.0
1
2
3
4
5
6
7
8 9
10
111213 14
15
1618
19
21
22
23
252628
29
30
33
3536
Figure V.11: Conventional GC chromatogram of butter FAMEs using a wax column: 30 m, 0.25 mm, 0.25 μm. For experimental conditions refer to text.
0.2 1.0 1.4Minutes
0 e 3
10 e 3
30 e 3
40 e 3
60 e 3
70 e 3
90 e 3
110 e 3
uV
1.80.6
1
2
3
4
5
6
7
8 9
10
111213 14
15
161819 21
22
23
2526
28
29
30
33
35 36
20 e 3
50 e 3
80 e 3
100 e 3
Figure V.12: Fast GC chromatogram of butter FAMEs using a wax column: 10 m, 0.1 mm, 0.1 μm. For experimental conditions refer to text.
Table V.4: Comparison of relative peak areas
for conventional and fast GC. Area reproducibility over 5 successive
runs was below 1 % RSD.
pling rate: 40 ms; filter time con-stant: 200 ms.
The parameters applied in the fastanalysis were as follows: injectionvolume: 0.2 μL (1:20 in hexane);split ratio: 1:200 (250 °C); tem-perature program: 50 °C to 250°C at 90.0 °C/min; head pressure:400 kPa at constant average linearvelocity; carrier gas: H2; [u]: 116.0cm/s; detector: FID (250 °C) H2:50 mL/min, air: 400 mL/min,make-up: 50 mL/min kPa (N2);sampling rate: 4 ms (250 Hz); filter time constant: 20 ms.
The correlation between fast andconventional GC data is verygood, as can be derived fromTable V.4, which reports relativepeak area percentages for bothapplications. For more details seereference [18].
1) C4 : 0 19) C15 : 02) C5 : 0 20) C15 : 13) C6 : 0 21) C16 : 0 iso4) C7 : 0 22) C16 : 05) C8 : 0 23) C16 : 16) C9 : 0 24) C17 : 0 anteiso7) C10 25) C17 : 0 iso8) C10 : 1 26) C17 : 09) C11 : 0 27) C17 : 110) C12 : 0 28) C18 : 0 iso11) C12 : 1 29) C18 : 012) C12 : 1 isomer 30) C18 : 1 �913) C13 : 0 31) C18 : 1 �714) C14 : 0 iso 32) C18 : 2 �?15) C14 : 0 33) C18 : 2 �616) C14 : 1 34) C18 : 2 �?17) C15 : 0 anteiso 35) C18 : 3 �318) C15 : 0 iso 36) C18 : 2 con.
Chromatography Volume 2 VI. APPLICATION IN THE PETROCHEMICAL FIELD
31
Kerosene analysis (GC-FID)
10 20 30
Chromatogram
Minutes
0 e 3100 e 3200 e 3300 e 3400 e 3500 e 3
700 e 3
1100 e 3
900 e 3
1300 e 31200 e 3
1400 e 3
600 e 3
1000 e 3
800 e 3
605040
uV
Figure VI.1: Conventional GC chromatogram recorded with a kerosene sample: RTX-1MS, 30 m, 0.25 mm ID, 0.25 μm
0.25 0.5 0.75
Minutes
0 e 350 e 3
200 e 3
300 e 3
400 e 3
500 e 3
700 e 3
350 e 3
150 e 3
550 e 3
450 e 3
650 e 3600 e 3
250 e 3
100 e 3
1.51.251.0
uV
750 e 3
2.252.01.75
Figure VI.2: Fast GC chromatogram relative to a kerosene sample: RTX-1MS, 10 m, 0.1 mm ID, 0.1 μm
GC analysis of kerosenemay be considered to bea very common quality
control procedure in the petro-chemical field.
An increase in sample through-put is of the highest interest inthis industrial sector. In FigureVI.1, the chromatogram recor dedwith a conventional 30 m columnis shown (temperature program:40 °C (2 min) at 2 °C/ min to 220 °C (3 min); carrier gas H2,linear velocity: 59 cm/s). Thetotal analysis time is about 72min. The predominant peaks, inan equally spaced order, arekerosene paraffines. The transi-tion to a fast method wasachieved using a RTX-1 MS, 10 m length, 0.1 mm ID, 0.1 μmstationary phase thickness col-umn. The result of the fast GCanalysis is shown in Figure VI.2.
In this case the analysis time isabout 2.3 min with a 31.3 foldspeed gain. The applied averagelinear velocity was 85.3 cm/s (H2as carrier gas) with a greatlyaccelerated temperature program.The split ratio was set at 800 :1with the gas saver function acti-vated (10 :1 after 1 min).
The temperature program was set from 40 °C up to 150 °C at 80 °C/ min, and up to 250 °C at70 °C/ min and finally up to 350 °C at 50 °C/min; the FID filter time constant and samplingfrequency were set to 10 ms and125 Hz respectively.
The chromatographic resolutionis very comparable with that ofthe conventional analysis. Thepeak widths of the narrowestpeaks are less than 0.5 s (FWHM).
01 Benzyl alcohol02 Limonene03 Linalool04 Methyl heptin carbonate05 Citronellol06 Citral (Neral)07 Cinnamic aldehyde08 Geraniol09 Citral (Geranial)10 Anisyl alcohol11 Hydroxy citronellal12 Cinnamyl alcohol13 Eugenol
14 Coumarin15 Iso-eugenol16 Methyl gamma ionone17 Lilial18 Amyl cinnamic aldehyde19 Lyral20 Amyl cinnamic alcohol21 Farnesol 122 Farnesol 223 Hexyl cinnamic aldehyde24 Benzyl benzoate25 Benzyl salycilate26 Benzyl cinnamate
1. Potential allergens in perfumes
In the 7th Amendment to theEuropean Cosmetics Directive,as published in the Official
Journal N° L 66 of the EuropeanUnion on March 11, 2003, 26 com-pounds are defined as potentialallergens. This regulation requiresthat the presence of these 26 fra-grance ingredients must be indi-cated on the label of finished cos-metic products if their concentra-tions exceed a threshold of 0.01 %for rinse-off and 0.001 % for leave-on products. These compounds arereferred to as potential allergens, asthey may or may not induce anallergic reaction. Among the 26materials indicated, two are naturalextracts (oak moss and tree moss)so the method as specified below isrestricted to the determination of24 volatile chemicals [19]. Thesecompounds are listed in TableVII.1.
It must be emphasized that thedetermination of these compo-nents, which is usually carried outusing quadrupole GC/MS equip-ment, is very complex and time-consuming. An example of such anapplication carried out with purestandard compounds is illustratedin Figure VII.1. The concentrationof each compound was about 400 ppm, with a total of 26 peaks(including isomers) separated on aCP SIL 5 50 m column with 0.25mm ID and 0.25 μm film thickness[temperature program: 50 °C
(1 min) to 210 °C at 2 °C/min, to280 °C at 10 °C/min (10 min). Carrier gas: He; average linearvelocity 34.4 cm/s; split ratio300 :1]. As the analysis time ismore than 75 minutes, the aim isto decrease this extensive GC runtime in order to increase the sam-ple throughput.
Figure VII.2 shows a fast GC/MSapplication carried out on a mix-ture of standard allergens with aRTX-5 10 m, 0.1 mm ID, 0.1 μmcolumn [temperature program: 70°C (1 min) to 180 °C at 25 °C/minto 280 °C (1 min) at 80 °C/min.Carrier gas: He; constant linearvelocity: 40 cm/s; split 300:1]. Inorder to evaluate the resolutionattained in both applications, twochromatographic expansions arecompared in Figure VII.3. As canbe seen, the degree of peak separa-tion is better in the fast application(citral and anisyl alcohol areresolved) while a speed gain of afactor of about 11 is observed.Peak FWHM values were on aver-age approx. 0.5 s; a mass range of30 to 350 amu and a 20 Hz acqui-sition rate were applied. The quali-ty of the resulting spectra, provid-ed by the GCMS-QP2010, wasvery high with similarities rangingbetween 94 and 98 (Wiley library).Linearity was measured between 4 and 400 ppm and a regressioncoefficient of 0.9995 was attainedwhich indicates excellent analytical
Table VII.1: Compounds defined by the IFRA as potential allergens
Figure VII.1: Conventional GC/MS chromatogram relative to an allergen standard mixture carried out with a CP SIL-5 50 m, 0.25 mm, 0.25 μm column
(x 1,000,000)
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.20
2.25
2.50
2.75
3.00
25.0 50.0 75.0
Figure VII.2: Fast GC/MS chromatogram relative to an allergen standard mixture carried out with a RTX-5 10 m, 0.1 mm, 0.1 μm column
(x 10,000,000)
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
2.0 3.0 4.0 5.0 6.0
33
Chromatography Volume 2 VII. APPLICATIONS IN THE FIELDS OF FLAVORS & FRAGRANCES
Chromatography Volume 2VII. APPLICATIONS IN THE FIELDS OF FLAVORS & FRAGRANCES
34
27.5 30.0
Minutes
0.250.50
1.25
1.75
2.25
2.75
3.75
2.00
1.00
3.00
2.50
3.503.25
1.50
0.75
35.032.5
TIC4.00
40.037.5
(x1,000,000)
4 / M
ethy
l Hep
tin C
arbo
nate
12 /
Cinn
amyl
Alc
ohol
11 /
Hyd
roxy
Citr
onel
lal
10 /
Ani
syl A
lcoh
ol9
/ Citr
al (G
eran
ial)
8 / G
eran
iol
7 / C
inna
mic
Ald
ehyd
e6
/ Citr
al (N
eral
)5
/ Citr
onel
lol
13 /
Euge
nol
3.00
Minutes
0.51.0
2.5
3.5
4.5
5.5
7.5
4.0
2.0
6.0
5.0
7.06.5
3.0
1.5
3.503.25
TIC8.0
4.003.75
(x10,000)
4 / M
ethy
l Hep
tin C
arbo
nate
12 /
Cinn
amyl
Alc
ohol
11 /
Hyd
roxy
Citr
onel
lal
10 /
Ani
syl A
lcoh
ol
9 / C
itral
(Ger
ania
l)
8 / G
eran
iol
7 / C
inna
mic
Ald
ehyd
e
6 / C
itral
(Ner
al)
5 / C
itron
ello
l
13 /
Euge
nol
CP SIL-5 50 m, 0.25 mm ID, 0.25 μm RTX-5 10 m, 0.1 mm ID, 0.1 μmSegment S1
Figure VII.3: Chromatographic expansions relative to Figure VII.1 and Figure VII.2
0 100 200
Limonene
R = 0.9995
Area (x 1,000,000)
con. (ppm)0.0
0.5
1.0
1.5
2.0
2.5
3.5
3.0
4.0
300 0 100 2000.0
1.0
2.0
5.0
3.0
4.0
Area (x 100,000)
Lyral
R = 0.9995
con. (ppm)
Figure VII.4: Calibration curves determined with fast GC/MS for limonene and lyral
2.0 3.0 4.0 5.0 6.0 7.6 7.6
Eau de Toilette
Minutes
TIC
16
17
19
25
Figure VII.5: Fast GC/MS analysis of allergens in a sample of Eau de Toilette. The peak numbers refer to Table VII.1.
precision. To illustrate, calibrationcurves for limonene and lyral areshown (Figure VII.4).
The method was then applied toreal-world samples. The TICresult of an Eau de Toilette sam-ple, diluted in acetone (1:100), isshown in Figure VII.5. The aller-gens found (concentrations referto the dilution) are methyl gammaionone (74.5 ppm), lilial (129.7ppm), lyral (99.4 ppm) and benzylsalicylate (96.6 ppm). The spectralsimilarities were between 93 and98 using a commercial MS library(Wiley).
35
Chromatography Volume 2 VII. APPLICATIONS IN THE FIELDS OF FLAVORS & FRAGRANCES
2. Quality control of flavors
In the flavor industry, routinequality control analyses areapplied to a high number of
samples. In this field, the analysisof essential oils is widespread andmay be considered as a complextask. Chromatograms relative togeranium and limette essential oilanalysis carried out on a DBWAX 10 m, 0.1 mm, 0.2 μm col-umn are reported in Figure VII.7and Figure VII.8.
The essential oils were diluted inethanol at a concentration of 5 %.The GC program was set from 40 °C (0.5 min) to 230 °C (at 50°C/min) with a 60 cm/s constantaverage linear velocity (hydro-gen). The injection volume was 1 μL with a split ratio of 400 :1.About 30 important flavor com-ponents were separated and iden-tified in less than 5 minutes.
0.5 1.0 1.5 2.0 2.5 3.0
cis-
beta
-Oci
men
e
Minutes3.5 4.0- 0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
uV (x 10,000)
Ger
anio
l
Citr
onel
lol
Ger
anyl
Ace
tate
Ger
anyl
For
mat
eCu
aiad
iene
+ C
itron
elly
le F
orm
ate
Ger
mac
rene
D
Ger
anyl
Isob
utyr
ate
Ner
ol
beta
-Car
yoph
ylen
eLi
nalo
ol
Lina
lool
oxid
e
Lim
onen
e
Isom
enth
one
Myr
cene
+ a
lpha
-Pha
ella
ndre
ne
Men
thon
eRose
Oxi
de
Para
cym
ene
Terp
inol
ene
alph
a-Pi
nene
tran
s-be
ta-O
cim
ene
gam
ma-
Terp
inen
e
4.5
Chromatogram
Rose
Oxi
de
Lina
lool
oxid
e
Ger
anyl
But
yrat
e
Citr
onel
lyle
Tig
late
Ger
anyl
Tig
late
Phen
ylet
hyle
Tig
late
10 g
amm
a-ep
i Eud
esm
ol
Figure VII.7: Fast GC/MS chromatogram (DB wax 10 m, 0.1 mm, 0.2 μm) of a geranium essential oil (5 % in ethanol)
0.5 1.0 1.5 2.0 3.0
cis-
beta
-Oci
men
e
Minutes3.5 4.0- 0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
uV (x 10,000)
Terp
in-1
-en-
4-ol
alph
a-Te
rpin
eol
Non
anal
Dec
anal
beta
-Car
yoph
ylle
ne
Lina
lool
alph
a-Te
rpin
ene
Lim
onen
e
Oct
anal
Myr
cene
+ a
lpha
-Pha
ella
ndre
ne
beta
-Phe
lland
rene
beta
-Pin
ene
Para
cym
ene
Terp
inol
ene
alph
a-Pi
nene tr
ans-
beta
-Oci
men
e
gam
ma-
Terp
inen
e
4.5
Chromatogram
Sabi
nene
Cine
ol 1
,8
gam
ma-
Terp
ineo
l
Cam
phen
e
Etha
nol
Figure VII.8: Fast GC/MS chromatogram (DB wax 10 m, 0.1 mm, 0.2 μm) of a limette essential oil (5 % in ethanol)
2.0 3.0 4.0 5.0 6.0 9.07.0
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
(x 10,000,000)
6-M
ethy
l-G
amm
a-Io
none
Lilia
l
Ger
anio
lCitr
onel
lol
Lina
lool
Lim
onen
e
8.0
TIC
alph
a-H
exyl
cinn
amic
al
dehy
de
0.5
0.0
Figure VII.6: Fast GC/MS TIC data relative to a perfume and obtained with a SPB-5 10 m, 0.1 mm, 0.1 μm column
The method was also applied to a higher concentrated perfume.Figure VII.6 shows the chromato-graphic result obtained with a 10% solution (acetone) of a perfumeusing a SPB 5 10 m, 0.1 mm, 0.1μm column (injection volume: 1 μL; split ratio: 500:1; average linear velocity: 50 cm/s; tempera-ture program: 50 °C (0.5 min) to200 °C at 20 °C/min, to 280 °C at50 °C/min; MS: scan range 41 to320 amu, 20 scans/s at 10,000amu/s scan rate). The identifica-tion of the potential allergen com-pounds was carried out by thefinder function of the GCMSsolu-tion software. In this case thewhole TIC is searched for alibrary match with the compoundsof interest. This is also shown inthe same figure. Seven allergens
were identified with a high simi-larity using the aforementionedcommercial library.
SI93969999989691
NameLimoneneLinanoolCitronellolGeraniol6-Methyl-Gamma-IononeLilialalpha-Hexylcinnamic aldehyde
Ret. Time2,6693,2194,184,3425,8256,2017,482
HCB18312853
Heptachlor52494435
Aldrine7470
p,p’-DDE101
99112
97110
77151149118153137138158187183156157180170198189194206209
Chromatography Volume 2VIII. APPLICATION IN THE ENVIRONMENTAL FIELD
38
3,00
3 4 5
Chromatogram
min
Figure VIII.1: GC-ECD chromatogram of 38 PCBs using a SPB5 10 m, 0.1 mm, 0.1 μm column. Concentrations are about 0.1 ppb.
Compound RT
TargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTargetTarget
Table VIII.1:List of PCBs contained
in the standard
PCB analysis with GC-ECD detection
The analysis of PCBs in dif-ferent matrices such as water or sludge is still
an important issue in the envi-ronmental field. These applica-tions are achieved by using eitherGC-ECD or NCI GC/MS. The results obtained by usingGC-ECD are shown in FigureVIII.1.
The chromatogram was attainedby injecting a mixture of 38 stan-dard PCBs (Table VIII.1). Thecolumn used was a SPB1, 10 m, 0.1 mm, 0.1 μm, with an
oven program of 90 °C (2 min) at40 °C/min to 290 °C and a con-stant average linear velocity of 50 cm/s (He). The injection vol-ume was 1 μL with a split ratioof 10 :1 (0.1 ppb for each com-pound). The ECD current wasset to 1 nA. The filter time con-stant and sampling frequencywere 20 ms and 100 Hz respec-tively. In order to avoid detectorband broadening, a make-up gasflow of 100 mL/min was applied.
4,5374,7464,9534,9575,0115,0655,1385,1615,2195,2385,2555,425,4375,5555,5775,6015,6485,6765,7035,7345,7455,8165,8695,9876,0726,1056,1226,1976,2236,3396,366,4266,4886,5486,6816,8397,0047,165
39
The reduction of analyticalcycle times in routineapplications has become
increasingly important in orderto increase productivity or toacquire quick and correct resultsin quality control processes.
The narrow-bore columnapproach is a very effective wayof decreasing GC run times
while maintaining a high degreeof peak resolution. The applica-tion of these methods is increa -sing in routine laboratory work,also due to improved column andGC instrumental technology. Infact, modern-day GC systemsfulfill the analytical requirementsof narrow-bore capillaries.
Literature
[1] Matter, L. (Ed.), Food and Environmental Analysis by Capillary Gas Chromatography, Hüthig, 1997
[2] van Es, A., High Speed Narrow-Bore Capillary Gas Chromatography,Hüthig, Heidelberg, 1992
[3] van Ysacker, P. G.; Janssen, H.-G. Snijders, H. M. J.; Cramers C. A.; J. High Resol. Chromatogr. 18 (1995) 397
[4] Broske, A. D. et al, Abstract A05 and A21, 20th International Symposium on Capillary Chromatography, Riva del Garda, Italy, May 26-29, 1998
[5] David, F. et al., Abstract P53, 20th International Symposium on Capillary Chromatography, Riva del Garda, Italy, May 26 -29, 1998
[6] Cook, W. S., Today’s Chemist at Work 1 (1996) 16[7] Sandra, P. et al., Experiments with fast Multicapillary GC, presented
at the 18th International Symposium on Capillary Chromatography,Riva del Garda, Italy, May 20-24, 1996
[8] van Lieshout, M. et al., J. Micro. Sep. 11 (2) (1999) 155[9] Schomburg, G., Gaschromatographie, Grundlagen, Praxis,
Kapillartechnik, VCH[10] Mondello, L. et al., J. Chrom. A, 1035 (2004) 237 [11] Grob, K., Classical Split/Splitless Injection, Hüthig [12] Hinshaw, J. V., LCGC (2002) vol 15, p. 152[13] Leoni, V. and D’Alessandro De Luca, E., Essenz. Deriv. Agrum.,
1978, 48, 39-50.[14] Kempe, G. and Baier, H.-U., Organophosphorous pesticides
determined in natural matrices with GC-FPD, GC-FTD and GCMS:medium and narrow-bore columns, presented at 25th InternationalSymposium on Capillary Chromatography, Riva del Garda, Italy, May 14-17, 2002.
[15] Mondello, L., 22th International Symposium on Capillary Chromatography, Riva del Garda, Italy May 2002
[16] Kondo, S. et al., Euroanalysis conference, 2004 [17] Han, J. J. and Yamane, T., Lipids, 1999, 34, 989-995 [18] Mondello, L., J. Microc. Sep. 12 (1) 41-47, 2000[19] www.ifraorg.org
Founded in 1875, Shimadzu Corporation, a leader in the development of ad vanced technologies, has a distinguished history of innovation built on the foundation of contributing to society through science and technology. We maintain a global network of sales, service, technical support and applications centers on six continents, and have established long-term relationships with a host of highly trained distributors located in over 100 countries.
For information about Shimadzu, and to contact your local office, please visit our website at www.shimadzu.eu
Shimadzu Europa GmbHAlbert-Hahn-Str. 6-10 · D-47269 DuisburgTel.: +49 - (0)203 - 76 87-0Fax: +49 - (0)203 - 76 66 [email protected]