amd analysis & technology ag

AMD Analysis & Technology AG

AMD Analysis & Technology Aktiengesellschaft • Königsberger Str. 1 • 27243 Harpstedt • Telefon (04244) 968736 • Telefax (04244) 968738 Vorstand: Diplom-Physiker Karl-Heinz Maurer (Vorstandsvorsitzender) • Aufsichtsratsvorsitzender: Dipl.-Oec. Mathias Maurer

Handelsregister HRB 141560 • Amtsgericht Oldenburg • USt-ID Nr. DE238516825 • Steuer-Nr. 2357/201/12096 • Zoll-Nr.: 6602150 Banken: OLB Harpstedt Kto. Nr. 246 450 20 00 (BLZ 280 233 25) ▪ IBAN: DE61280200502464502000 ▪ SWIFT OLBO DE H2

Technical Note 120604 Author: Karl-Heinz Maurer AMD Mini-QuAS3AR, a new API-MS Mass Spectrometer A) Introduction Previous Application Notes describe some results obtained with the existing API-MS Evaluation Model: Application Note 120329 (ESI-MS Elemental Trace Analysis of Alkali metals in water); Application Note 120419 (APCI-MS Trace Analysis of volatile organic compounds in ambient air); and Application Note 120427 (GD-MS Trace Analysis of volatile organic compounds in ambient air). Supplementary, we describe here in short-form some technical aspects of the miniaturized API-MS System, considered to be the basis for dedicated product versions in above mentioned application areas. B) Short-form Description of a unique System

Fig. 1 Evaluation Model of the new bench-top Mass Spectrometer AMD Mini-QuAS3AR, equipped with an Atmospheric Pressure Ionization (API) Interface

70 cm

Magnetic Analyzer Electric Sector Analyzer API Interface API Housing Multi-Stage Turbo Pumping System

, Technical Note 120604

Analysis & Technology Aktiengesellschaft • Königsberger Str. 1 • 27243 Harpstedt • Telefon (04244) 968736 • Telefax (04244) 968738 Vorstand: Diplom-Physiker Karl-Heinz Maurer (Vorstandsvorsitzender) • Aufsichtsratsvorsitzender: Dipl.-Oec. Mathias Maurer

Handelsregister HRB 141560 • Amtsgericht Oldenburg • Ust-ID Nr. DE238516825 • Steuer-Nr. 2357/201/12096 • Zoll-Nr.: 6602150 Banken: OLB Harpstedt Kto. Nr. 246 450 20 00 (BLZ 280 233 25) ▪ IBAN: DE61280200502464502000 ▪ SWIFT OLBO DE H2

Fig. 2 Screen shot of the AMD Multi-Instrument-Control System (MICS) and AMD data system The experimental version of the AMD Mini-QuAS3AR is a double focusing mass spectrometer based on a Mattauch-Herzog ion optical design with a straight focal plane suitable for the incorporation of an array detector for simultaneous ion detection in a wide mass range. The dedicated and miniaturized system is based on the original AMD QuAS3AR Technology. Emphasis regarding the API interface and the analyzer design has been placed to the analysis of low mass ions. The system incorporates the significant advantages of magnetic sector mass spectrometers for qualitative and quantitative analyses of low mass ions regarding peak shape, resolution and abundance sensitivity. The API-MS interface and the mass analyzer are integrated with a multi-stage turbo pumping system. C) Technical Data Summary 1 Analyzer: Magnetic sector analyzer system in double focusing EB configuration with straight focal plane 2 Resolution: Point detector: 30 - 300 (10 % valley) selectable 3 Mass range: 1 - 400 Dalton at 2 kV accelerating voltage 4 Scan range: B-scan: full mass range V/E-scan: 0.3 mass decade (factor 2) 5 Scan speed: B-scan: max. 0.5 sec/mass decade

V/E-scan: max. 30 msec/scan 6 Limit of Detection: ppb – low ppt range (depending on ionization method and sample) 7 Polarity: Kation or Anion detection 8 Instrument Control: Multi-Tasking Instrument-Control-System (AMD MICS) for processor

controlled system functions 9 Data System: AMD Data system (version 2012); PC, state of the art technology, Windows

operating system 10 Ionization Methods: ESI/APCI,Glow Discharge with API Interface optimized for low mass ions

6Li

7Li




Application Note 120329 Author: Karl-Heinz Maurer ESI-MS Elemental Trace Analysis of Alkali metals in water

A) Introduction Elemental analysis is usually a domain of analytical methods like AAS, ICP-OES and ICP-MS. A standard method for multi element trace analysis using mass spectrometry is the production of Kations by ICP techniques and the application of a corresponding interface to the MS analyzer. For mass spectrometric analysis of polar and non volatile molecules the method of choice are API-MS techniques. Ion production takes place by ESI or APCI methodologies and dedicated interfaces to the MS analyzer are applied. Usually, the common API-MS interfaces are not suitable or at least very much restricted regarding the analysis of low mass ions. We were interested to evaluate the capabilities of a specific ESI-MS system, especially for direct analysis of low atomic number elements.

B) Summary ESI-MS analyses have been performed with the experimental version of a new bench-top double focusing mass spectrometer. The dedicated and miniaturized system is based on the original AMD QuAS3AR Technology. Emphasis has been placed to the analysis of low mass ions and logically, the analysis of Lithium as the lowest element of the alkali metals in the first main group of the periodic system got some priority. Within the scope of the reported quantification methods no official certified procedures or U.S. EPA methods were applied, since the aim was to achieve orientating results for possible future applications. The ESI-MS interface and the mass analyzer are integrated with a multi-stage turbo pumping system. It has been demonstrated that the special features of a small double focusing mass spectrometer regarding peak shape, mass resolution and abundance sensitivity are of significant advantage for the qualitative and quantitative analysis of ions in the low mass range. Quantitative Trace Analyses of Lithium in aqueous solutions and in mineral water as an exemplary real life example were performed. Detection limits in the low ppt range were achieved for Lithium. The influence of scan methodologies including raw data accumulation and selected ion monitoring techniques on the quality of the results and analysis time are described. The response factors of other alkali metals have been determined, approximately and have been found to be similar for these elements. The response factors for the alkaline earth metals and other elements are expected to be lower and require more investigation in order to conclude a possible usefulness of ESI-MS for quantitative analyses. The results of our evaluations of ESI-MS applications so far dedicated to the alkali metals indicate that at least the quantitative analysis of Lithium in aqueous samples can be performed, success-fully and limits of detection in the low ppt range can be reached. This method does not require any specific sample preparation and no chromatographic separation.

, Application Note 120329



C) Methodology

C.1 Mass spectrometer

The mass analyzer was a miniaturized dedicated version of the AMD QuAS3AR (Quasar070206.pdf) double focusing system, equipped with an API Interface in ESI mode (ESI-MS system). System details are not reported here since an evaluation model was used for the experiments. Fig. 1 Scheme of the ion optics and physical size of the bench-top double focusing MS analyzer

Fig. 2 CAD Design of the Miniaturized AMD QuAS3AR API-MS system

Miniaturized AMD QuAS3AR

Magnetic Sector Analyzer

Electric Sector

Entrance slit

Array Detector (not yet implemented)

48 cm

25 cm

The miniaturized AMD QuAS3AR is based on the original AMD QuAS3AR Technology as a multifunctional high performance system and previously described in a system description: SD040617_ AMD Quasar.pdf. The new current miniaturized version is downsized significantly, dedicated to low mass resolution and API-Interfacing only. However, it incorporates the significant advantages of magnetic sector mass spectrometers for low mass ions regarding resolution and abundance sensitivity

Multi stage turbo pumping system.

Magnetic sector analyzer




C.2 ESI-MS Interface As a dedicated API-MS bench-top system the interface and the mass analyzer are mounted on a multistage turbo pumping system. The system is fully optimized for highest transmission of low mass ions through the interface to the MS detector and in so far specifically dedicated to trace analysis of Lithium, the lowest atomic number element of the alkali metals.

Fig. 3 ESI-MS interface front view of the miniaturized API-MS Evaluation Model

C.3 Measurement procedure for Lithium analysis

Fig. 4 ESI-MS spectrum of the Lithium Isotopes at a concentration of 10-8 Mol/L

The API ion source of the miniaturized AMD QuAS3AR dedicated for ESI-MS applications consists of an API room including nebulizer assisted sprayer and hot air pipes for desolvatisation assistance. The multistage vacuum interface is integrated with the analyzer multistage turbo pumping system. A continuous eluent ESI spray of 85 µL/min was maintained by a syringe pump and assisted by a nebulizer gas flow of 1.2 l/min.

Analysis Details C.3 Sample: 10-8 molar Lithium standard in aqueous solution, diluted 1:2 with ultra pure Methanol Scan method: magnet scan, Scan duration: 850 msec Measuring time per peak: 40 msec Data acquisition: raw data accumulation (200 spectra) Detector: post acceleration and Channeltron type SEM

7Li

6Li

Total Ion Current





Fig. 6 Details of the Lithium Isotopes at a concentration of 10-8 Mol/L

Analysis Details C.3 cont’d Mass resolution: 50 at 10% valley with fixed slits Data System: AMD Version 12.01.1 Analysis time: Since no chromatography was used, data were recorded over a time of about 300 sec and accumulated, accordingly

Analysis Details C.3 cont’d Detection Limits: Both Lithium Isotopes are well detected at a concen-tration of c= 7 x 10-8 g/L or c = 7 x 10-11 = 70 ppt The 6Li Isotope with a low abundance of 7.5 % yields still a S/N ratio of about 4:1

7Li

6Li

7Li

6Li

Total Ion Current




Fig. 7 S/N for the 7Li Isotope at a concentration of 10-8 Mol/L (70 ppt)

Fig. 8 Background (Blank) signal for Lithium determined after a number of measurements

Analysis Details C.3 cont’d The major isotope 7Li yields a Signal to Noise ratio of about S/N= 1000 µV/25 µV or S/N = 40:1 If the limit of detection is reached at a S/N = 4:1 a value LODLithium = 7 ppt can be derived, if a linear calibration function is as-sumed and the background (blank) signal may be neglected or small enough for deduction

Analysis Details C.3 cont’d Background signal The limit of detection (LOD) may be influenced by a background signal. The 7Li blank signal is here about 330 µV which would have to be deducted from the signal in Fig.7, resulting in a LOD of about 10 ppt. For precise quantification a calibration curve has to be established

7Li

7Li




D) Results

D.1 Lithium Analysis in Mineral Water


Fig. 10 Details of the Lithium Isotopes at a concentration of 10-7 Mol/L

Analysis Details D.1 In preparation of a quanti-tative analysis of Lithium in a Mineral Water sample a 10-7 molar (0.7 ppb) Lithium standard has been measure-ed. Spectrum a has been ob-tained with an accumulation time of 20 sec and spectrum b with a longer accumulation of about 300 sec. It is obvious that the S/N ratio and the LOD improve at longer accumulation times

Analysis Details D.1 cont’d Both Lithium Isotopes yield intensities which allow a precise determination of the isotopic ratio at a concen-tration of 0.7 ppb This will be of importance if isotope dilution techniques will be applied for accurate and precise quantification. This technique was not avail-able within the scope of this application note.

a

b

7Li

7Li

6Li




D.2 Analytical Result of Lithium Analyses in Vilsa Medium Mineral Water

Fig. 11 ESI-MS spectrum of the Lithium Isotopes of Vilsa Medium Mineral Water

Fig. 12 Details of the Lithium Isotopes of the Vilsa Medium Mineral Water

Analysis Details D.2 A sample of Vilsa Medium Mineral Water has been analyzed for the Lithium content. The analysis was based on a direct comparison of the Lithium ion intensities with the 0.7 ppb standard. In so far the result (see below) may be considered to be semi-quantitative

Analysis Details D.2 cont’d The intensities and the ratio of the Lithium Isotopes in the Vilsa Medium Mineral Water sample have been determined, precisely The details of the peak shapes and the very high S/N ratio indicate the quality potential for a quantitative data evaluation (see below), typical for a magnetic sector mass spectrometer




Fig. 13 Isotopic ratio determination of the Lithium Isotopes of the Vilsa Medium Mineral Water sample

Fig. 14 ESI-MS SIM measurement of the Lithium Isotopes of Vilsa Medium Mineral Water

Analysis Details D.2 cont’d The Lithium isotope ratio has been determined by evaluation of the peak areas (blue lines in the figure) The theoretical ratio is 6Li/7Li = 0.0813 The measured value is 6Li/7Li = 0.0828 The relative deviation is ∆ = 1.84%

Analysis Details D.2 cont’d While all above measure-ments have been per-formed in scan mode, another measurement of the Vilsa Medium Mineral Water was performed in SIM mode. Due to longer measuring time per peak, this results in shorter analysis time of about 60 sec for the same S/N ratio and precision result. This may be of importance for automated multi-sample analysis

SIM Method

7Li 6Li




D.3 Analysis of Alkali Metals in aqueous solution

Fig. 15 ESI-MS spectrum of Alkali Metals in a 10-5 molar standard mixture

D.4 Analysis of Alkali Metals Vilsa Medium Mineral Water

Fig. 16 ESI-MS spectrum of Alkali Metals in of Vilsa Medium Mineral Water

Analysis Details D.3 The response of three Alkali Metals for ESI-MS has been tested in scan mode The result for a 10-5 molar standard mixture is shown in this figure. The response factors are similar for all elements, which is also valid for 133Cs which has been measured but is not shown in this spectrum. Background ions of organic nature occur, too and will especially influence the detection limits for elements with low response factors.

7Li 23Na 39K

Analysis Details D.4 This survey spectrum of Vilsa Medium Mineral Water has not been evalu-ated, quantitatively. There is some background contribution to the 39K ion while the 7Li intensity is in accord to the quantitative measurements above and the strong 23Na intensity corresponds to an estima-ted concentration of about 1 ppm. This is lower than indicated on the bottle but corresponds with the label. “natriumarm”.

7Li

23Na

39K




Application Note 120419 Author: Karl-Heinz Maurer APCI-MS Trace Analysis of volatile organic compounds in ambient air

A) Introduction Trace analysis of volatile organic compounds (VOCs) by mass spectrometry (MS) is usually performed by Electron impact (EI) or classical Chemical Ionization techniques (CI). In these cases the matrix gas is introduced by a heated capillary interface and the matrix gas amount used for MS analysis is in the order of 0.3 -1 ml/min. Special technologies as proton transfer reaction mass spectrometry (PTR-MS) and related selected ion-flow-tube mass spectrometry (SIFT-MS) play a role, too. For mass spectrometric analysis of polar and non volatile molecules in liquid matrices the established method of choice are API-MS techniques with dedicated interfaces. We have recently reported (Application Note 120329) about a specific ESI-MS system, suitable for the direct analysis of low atomic number elements. In addition we were interested to evaluate the capabilities of this system for the direct analysis of volatile organic compounds in ambient air by the application of Atmospheric Pressure Corona Discharge Ionization (APCDI). The ambient air acts as reagent gas in this case, therefore the term APCI may be used. No GC separation and purge and trap techniques have been applied.

B) Summary Analyses have been performed with the experimental version of a new bench-top double focusing mass spectrometer. The dedicated and miniaturized system is based on the original AMD QuAS3AR Technology. Emphasis has been placed to the analysis of volatile organic compounds in the gas phase, logically producing mass spectra of low molecular weight compounds. Within the scope of the reported semi-quantification methods no official certified procedures were applied, since the aim here was to achieve orientating results for possible future applications. The existing API Interface useable for ESI-MS or APCI-MS and the mass analyzer are integrated with a multi-stage turbo pumping system. It has been demonstrated again that the special features of a small double focusing mass spectrometer regarding peak shape, mass resolution and abundance sensitivity are of significant advantage for the qualitative and quantitative analysis of ions in the low mass range. Semi-Quantitative Trace Analyses of a number of VOCs (typically used as solvents) in ambient air were performed, exemplarily. Limits of detection (LOD) in the upper ppb range were achieved for a number of these compounds in full scan mode. SIM techniques which would improve sensitivity and extend the LOD to the low ppb range have not been applied because specificity was considered to be more important in this case. The results depend on the ionization efficiency and the production of quasi-molecular ions, respectively on the low number of key fragments which is typical for Chemical Ionization. The influence of scan methodologies including raw data accumu-lation techniques on the quality of the results and analysis time are also described. The achieved results of APCI-MS applications on VOCs in ambient air indicate the usefulness of the method for trace analysis in gaseous matrices. This may be of interest for environmental or process analysis in a wide range of concentrations. Dependent on the complexity of the gaseous matrix the method may be applied without chromatographic separation (GC). In this context it is of specificity advantage that either mainly quasi-molecular ions or only a few intense key fragment ions are formed.




C) Methodology


The mass analyzer was a miniaturized dedicated version of the AMD QuAS3AR (Quasar070206.pdf) double focusing system, equipped with an API Interface in APCI mode (APCI-MS system). System details are not reported here since an evaluation model was used for the experiments. Fig. 1 Scheme of the ion optics and physical size of the bench-top double focusing MS analyzer




Electric Sector

Entrance slit


48 cm

25 cm

The miniaturized AMD QuAS3AR is based on the original AMD QuAS3AR Technology as a multifunctional high performance system and previously described in a system description: SD040617_ AMD Quasar.pdf. The new current miniaturized version is downsized significantly, dedicated to low mass resolution and API-Interfacing only. However, it incorporates the significant advantages of magnetic sector mass spectrometers for low mass ions regarding resolution and abundance sensitivity






C.2 API-MS Interface As a dedicated API-MS (APCI or ESI) bench-top system the interface and the mass analyzer are mounted on a multistage turbo pumping system. The system is fully optimized for highest trans-mission of low mass ions through the interface to the MS detector and in so far relevant for the applications reported here, specifically dedicated to trace analysis of VOCs in ambient air.

Fig. 3 APCI (ESI)-MS interface front view of the miniaturized API-MS Evaluation Model

C.3 Measurement procedure for VOC analysis (example: Ethyl acetate) in ambient air

Fig. 4 APCI-MS corona discharge mass spectrum of Ethyl acetate in ambient air

The API ion source of the miniaturized AMD QuAS3AR dedicated for APCI (ESI)-MS applications consists of an API room including nebulizer assisted sprayer and hot air pipes for desol-vation assistance in ESI mode. which are not used for the APCI experiments described here. A corona discharge needle is mounted in front of the counter electrode orifice. The multi-stage vacuum interface is integrated with the analyzer multi-stage turbo pumping system.

Analysis Details C.3 Matrix: ambient air Sample: Ethyl acetate Sample amount: 1 µl Sample introduction: Syringe injection in API room (2 l ambient air volume) at 40 0C air temperature Ionization: Corona discharge Uc ≈ +3000 V Sample concentration after injection and evaporation (approximate calculation): c = 9 x10-4g / 2.5 g = 3.6 10-4 c = 360 ppm API Interface air intake: 0.4 l/min from API room

ESI Parts

Corona needle

C2H5O2 (C4H8O2 +H)+

Ion chromatogram of key fragment (Decline Curve)

Air exchange time




Fig.5 APCI-MS peak display of corona discharge mass spectrum of Ethyl acetate in ambient air

Fig. 6 APCI-MS corona discharge mass spectrum (single scan) of Ethyl acetate (diluted in Methanol) in ambient air. Methanol yield is about 4 times less than for Ethyl acetate

Analysis Details C.3 cont’d Scan method: magnet scan, Mass resolution: 70 at 10% valley with fixed slits Scan Speed: 1 sec/dec Measuring time per peak: 12 msec Data acquisition: single scan or raw data accumulation Detector: post acceleration and Channeltron type SEM Data System: AMD Version 12.01.1 Analysis time: Data were recorded over the a time of about 1400 sec until the sample concentration in the API room was below detection limit (see below)

Analysis Details C.3 cont’d Limit of Detection (LOD) Matrix: ambient air Sample: Ethyl acetate diluted in Methanol (1:100) Sample amount: 10 nl in 1 µl Methanol Sample introduction: Injection in API room as before Ionization: Corona discharge as before Sample concentration after injection and evaporation (approximate calculation): c = 9 x10-6g / 2.5 g = 3.6 10-6 c = 3.6 ppm API Interface air intake: as before

Ion chromatogram of key fragment at mass 61 (Decline Curve)

Peak width:12 msec

Peak shape at resolution 70 (10% valley)

Ion chromatogram of quasi molecular ion of methanol (Decline Curve)

Ion chromatogram of key fragment of Ethyl acetate at mass 61 (Decline Curve)

MeOH

Ethyl acetate




Fig. 7 APCI-MS display of key fragment ion of corona discharge mass spectrum of Ethyl acetate in ambient air at limit of detection

C.3.1 Short-form summary of the applied methodology and the result for APCI-MS analysis of Ethyl acetate (Fig. 4 -7) in ambient air The compound has been used as an example to demonstrate the measurement procedures for APCI-MS trace analysis of a volatile organic compound in ambient air. The rudimentary method used here should give some orientating results for the application of a possible analytical procedure under more sophisticated sample preparation and sample introduction methodology. The counter electrode/nozzle part of the API interface was kept at a temperature of 150 0C resulting in an ambient air temperature within the API room of about 40 0C. The volatile compound was injected by a syringe directly into the air of the API room. As described in figures 4 and 5 above an amount of 1 µl Ethyl acetate was injected and evaporated immediately. The API interface air intake of 0.4 l/min diluted the evaporated sample in the ambient air of the API room (2 l volume) by air continuous exchange from outside. The immediate sample concentration was calculated (see fig. 4) by the ratio of evaporated sample quantity to the air quantity in the API room (temperature corrected). According to the decline function about 40% of the sample at higher concentrations may be adsorbed on the inner walls and is desorbing, slowly. In the low ppm or even ppb range a single scan may be sufficient for identification of the compound (see fig 6) in the air. The sensitivity of the method allows detection limits in the low ppb range, if SIM techniques (longer measuring times per ion, higher number of ions but short total analysis time) will be used. However, the individual background signals are the limiting factors. Routine applications will require a controlled introduction (gas pumping) of the air matrix containing the analyte gas sample into the heated counter electrode/nozzle zone via a heated inlet line. The use of external heated volumes for the sample and the clean flushing gas as well as corresponding switch valves will be mandatory for efficient and precise analyses.

Analysis Details C.3 cont’d Limit of Detection (LOD) Matrix: ambient air Sample: Ethyl acetate diluted in Methanol 1:500 Sample amount: 2 nl in 1 µl Methanol Sample introduction: as before Ionization: as before Sample concentration after injection and evaporation (approximate calculation): c = 1.8 x10-6g /2.5 g = 0.72 10-6 c = 720 ppb API Interface air intake as before

Ion chromatogram of key fragment of Ethyl acetate at mass 61 (Decline Curve)

background

sample




D. APCI-MS trace analysis of Acetonitril (ACN) in ambient air

Fig. 8 APCI-MS corona discharge mass spectrum (single scan) of Acetonitril (diluted in Methanol) in ambient air.

Fig. 9 Ion chromatograms of Quasi-molecular ion and background signal (S/N ratio = 3:1) of Acetonitril in ambient air at limit of detection (LOD ≈ 180 ppb).

Analysis Details D. Limit of Detection (LOD) Matrix: ambient air Sample: Acetonitril diluted in Methanol 1:1000 Sample amount: 1 nl in 1 µl Methanol Sample introduction: as before Ionization: as before Sample concentration (definition as before) c = 9 x10-7g/2.5g = 3.6 10-7 c = 360 ppb API Interface air intake as before

Ion chromatogram quasi molecular ion of Acetonitril at mass 42 (incl. background)

background

sample

Ion chromatogram quasi molecular ion of Acetonitril at mass 42 (Decline Curve)

Ion chromatogram of mass 42 (background signal)

Analysis Details D. cont’d Limit of Detection (LOD) Matrix: ambient air Sample: Acetonitril diluted in Methanol 1:1000 Sample amount: 1 nl in 1 µl Methanol Sample introduction: as before Ionization: as before Sample concentration (definition as before) c = 9 x10-7g/2.5g = 3.6 10-7 c = 360 ppb API Interface air intake as before

(M+H)+

Noise level ≈ 3 mV




E. APCI-MS trace analysis of Acetone in ambient air

Fig. 10 APCI-MS corona discharge mass spectrum (single scan) of Acetone in ambient air.

Fig. 11 APCI-MS corona discharge mass spectrum (single scan) of Acetone in ambient air including background signals at quasi molecular ion

Analysis Details E. Matrix: ambient air Sample: Acetone Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 8 x10-4g/2.5g = 3.2 10-4 c = 320 ppm Dynamic range: see below

Analysis Details E. cont’d Matrix: ambient air Sample: Acetone Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 8 x10-4g/2.5g = 3.2 10-4 c = 320 ppm Dynamic range: 1:200 (2 mV peak detectable) LOD: about 3 ppm (high background signal)

Ion chromatogram quasi molecular ion of Acetone at mass 59 (Decline Curve)

(M+H)+

Peak display quasi molecular ion

background signal




F. APCI-MS trace analysis of Methanol in ambient air

Fig. 12 APCI-MS corona discharge mass spectra of Methanol in ambient air G. APCI-MS trace analysis of Gasoline in ambient air

Fig.13 APCI-MS corona discharge mass spectra of Gasoline in ambient air. The spectra indicate the mixture of major acyclic hydrocarbons. The method may be useable in this or similar cases for fingerprint analyses but requires further investigations

Analysis Details F. Matrix: ambient air Sample: Methanol Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 8 x10-4g/2.5g = 3.2 10-4 c = 320 ppm Dynamic range: 1:100 LOD estimated: < 3 ppm

Analysis Details G. Matrix: ambient air Sample: Gasoline Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 8 x10-4g/2.5g = 3.2 10-4 c = 320 ppm Dynamic range: 1:200 LOD: Not to be defined for one component


Ion chromatogram of quasi-molecular ion at mass 33 (Decline Curve)

(M+H)+


single spectrum

accumulated spectra




H. APCI-MS trace analysis of volatile halogen compounds in ambient air using negative ion detection

H.1 APCI-MS trace analysis of Trifluor acedic acid in ambient air

Fig. 14 APCI-MS corona discharge negative ion mass spectra of Trifluor acedic acid in ambient air

H.2 APCI-MS trace analysis of Dichlormethane in ambient air

Fig. 15 APCI-MS corona discharge negative ion mass spectra of Dichlormethane in ambient air.

Analysis Details H.1 Matrix: ambient air Sample:Trifluor acedic acid Sample amount: 1 µl Sample introduction: as before Ionization:Corona discharge Uc ≈ - 3000 V for negative ion production API Interface air intake as before Sample concentration (definition as before) c = 1.5 x10-3g/2.5g = 6 10-4 c = 600 ppm LOD: 100 ppm (estimated)

F- CF3- (M-H)

-

no background signals at key sample ion positions

(M-H) -

Cl35 -

Cl37 -

Analysis Details H.2 Matrix: ambient air Sample: Dichlormethane Sample amount: 1 µl Sample introduction: as before Ionization:Corona discharge Uc ≈ - 3000 V for negative ion production API Interface air intake as before Sample concentration (definition as before) c = 1.3 x10-3g/2.5g = 5 10-4 c = 500 ppm LOD: 150 ppm (estimated)




H.3 APCI-MS trace analysis of Chloroform in ambient air

Fig. 16 APCI-MS corona discharge negative ion mass spectra of Chloroform in ambient air.

Fig. 17 APCI-MS detailed view on Chlorine Isotopes of corona discharge negative ion mass spectrum of Chloroform in ambient air.

Analysis Details H.3 Matrix: ambient air Sample: Chloroform Sample amount: 1 µl Sample introduction: as before Ionization:Corona discharge Uc ≈ - 3000 V for negative ion production API Interface air intake as before Sample concentration (definition as before) c = 1.5 x10-3g/2.5g = 6 10-4 c = 600 ppm

accumulated sample spectra

accumulated background spectra

Chlorine Isotopes

Cl35 -

Cl37 -

Analysis Details H.3 cont’d Matrix: ambient air Sample: Chloroform Sample amount: 1 µl Sample introduction: as before Ionization:Corona discharge Uc ≈ - 3000 V for negative ion production API Interface air intake as before Sample concentration (definition as before) c = 1.5 x10-3g/2.5g = 6 10-4 c = 600 ppm LOD: 100 ppm (estimated)




H. 4 Short-form summary of APCI-MS trace analysis of volatile halogen compounds in ambient air using negative ion detection

The results for demonstration of the applicability of negative ion recording to volatile halogen compounds were achieved under the same rudimentary method as mentioned before for the VOC analyses in positive ion recording mode. Further investigations for clarification of improved limits of detection were beyond the scope of this application note. While for negative ion detection the background is not the limiting factor, the absolute sensitivity achieved here for the method requires further investigations regarding possible improvements. Otherwise the remarks for routine applications mentioned under C.3.1 above are applicable, too.




Application Note 120427 Author: Karl-Heinz Maurer GD-MS Trace Analysis of volatile organic compounds in ambient air

A) Introduction Trace analysis of volatile organic compounds (VOCs) by mass spectrometry (MS) is usually performed by Electron impact (EI) or classical Chemical Ionization techniques (CI). In these cases the matrix gas is introduced by a heated capillary interface and the matrix gas amount used for MS analysis is in the order of 0.3 -1 ml/min. Special technologies as proton transfer reaction mass spectrometry (PTR-MS) and related selected ion-flow-tube mass spectrometry (SIFT-MS) play a role, too. For mass spectrometric analysis of polar and non volatile molecules in liquid matrices the established method of choice are API-MS techniques with dedicated interfaces. We have recently reported (Application Note 120329) about a specific ESI-MS system, suitable for the direct analysis of low atomic number elements. We also reported (Application Note 120419) about this API-MS system used for the direct analysis of volatile organic compounds in ambient air by the application of Atmospheric Pressure Corona Discharge Ionization (APCI). In addition we were interested to evaluate the same system for the direct analysis of volatile organic compounds in ambient air by the application Glow Discharge Ionization (GD-MS). No GC separation and purge and trap or selective membrane enrichment techniques have been applied.

B) Summary As previously reported analyses have been performed with the experimental version of a new bench-top double focusing mass spectrometer of Mattauch-Herzog type. The dedicated and miniaturized system is based on the original AMD QuAS3AR Technology. Emphasis has been placed to the analysis of volatile organic compounds in the gas phase, logically producing mass spectra of low molecular weight compounds. Within the scope of the reported semi-quantification methods no official certified procedures were applied, since the aim here was to achieve orientating results for possible future applications. The existing API Interface useable for ESI-MS, APCI-MS, GD-MS and the mass analyzer are integrated with a multi-stage turbo pumping system. It has been demonstrated again that the special features of a small double focusing mass spectrometer regarding peak shape, mass resolution and abundance sensitivity are of significant advantage for the qualitative and quantitative analysis of ions in the low mass range. Semi-Quantitative Trace Analyses of a number of VOCs (typically used as solvents) in ambient air were performed, exemplarily. Limits of detection (LOD) in the upper ppb range were achieved for a number of these compounds in full scan mode. SIM techniques extended the LOD to the low ppb range. The results depend on the ionization efficiency and the production of quasi-molecular ions, respectively on the low number of key fragments which is typical for a Chemical Ionization process. The achieved results here using GD-MS applications on VOCs in ambient air indicate the usefulness of the method for trace analysis in gaseous matrices. This may be of interest for environmental and process analysis as well as for medical research applications in a wide range of concentrations. Depending on the complexity of the gaseous matrix the method may be applied without chromatographic separation (GC). In this context it is of specificity advantage that either mainly quasi-molecular ions or only a few intense key fragment ions are formed.




C) Methodology


The mass analyzer was a miniaturized dedicated version of the AMD QuAS3AR (Quasar070206.pdf) double focusing system, equipped with a universal API Interface in GD mode (GD-MS system). System details are not reported here since an evaluation model was used for the experiments. Fig. 1 Scheme of the ion optics and physical size of the bench-top double focusing MS analyzer




Electric Sector

Entrance slit


48 cm

25 cm

The miniaturized AMD QuAS3AR is based on the original AMD QuAS3AR Technology as a multifunctional high performance system and previously described in a system description: SD040617_ AMD Quasar.pdf. The new current miniaturized version is downsized significantly, dedicated to low mass resolution and API-Interfacing, currently However, it incorporates the significant advantages of magnetic sector mass spectrometers for low mass ions regarding resolution and abundance sensitivity






C.2 API-MS Interface

As a dedicated API-MS (APCI or ESI) bench-top system the interface and the mass analyzer are mounted on a multistage turbo pumping system. The system is fully optimized for highest trans-mission of low mass ions through the interface to the MS detector and in so far relevant for the applications reported here in GD-MS mode for trace analysis of VOCs in ambient air.

Fig. 3 API-MS interface in GD-MS mode for multiple applications; front view of the miniaturized API-MS Evaluation Model

C.3 GD-MS measurement procedure for VOC analysis in ambient air

Fig. 4 GD-MS mass spectra of Acetonitril and background in ambient air. Water cluster and oxygen matrix ions form the background in all spectra (see also below)

The API ion source of the miniaturized AMD QuAS3AR dedicated for APCI (ESI)-MS applications consists of an API room including nebulizer assisted sprayer and hot air pipes for desol-vation assistance in ESI mode. which are not used for the GD-MS experiments described here. Also the corona discharge needle mounted in front of the counter electrode orifice was not used. The multi-stage vacuum interface is integrated with the analyzer multi-stage turbo pumping system. Glow discharge was initiated in the first stage at a pressure of 1 mbar

Analysis Details C.3 Matrix: ambient air Sample: Acetonitril Sample amount: 1 µl Sample introduction: Syringe injection in API room (2 l ambient air volume) at 20 0C air temperature Ionization: Glow Discharge at 1 mbar in stage 1 of the API interface Sample concentration: after injection and evaporation c= 9 x10-4g / 2.6 g = 3.46 10-4 c = 346 ppm API Interface air intake: 0.4 l/min from API room

ESI Parts

Corona needle ESI and Corona parts not used for GD applications

Ion chromatogram of quasi molecular ion of Acetonitril at mass 42 (decline curve)

(M+H)+

background

O2+

(H2O+H)+ ((H2O)2+H)+

sample




Fig. 5 GD-MS partial mass spectrum of Acetonitril in ambient air

Fig. 6 GD-MS ion chromatogram of quasi-molecular ion of Acetonitril in ambient air

Analysis Details C.3 cont’d Scan method: magnet scan, Mass resolution: 60 at 10% valley with fixed slits Scan Speed: 2 sec/dec Measuring time per peak: 18 msec Data acquisition: single scan or raw data accumulation Detector: post acceleration and Channeltron type SEM Data System: AMD Version 12.01.1 Analysis time: Data were recorded over the a time of about 1300 sec until the sample concentration in the API room was at detection limit

Analysis Details C.3 cont’d Limit of Detection (LOD) Matrix: ambient air Sample: Acetonitril Sample amount: 1 µl Sample introduction: Injection in API room as before Ionization: Glow discharge as before Sample concentration after injection and evaporation c= 9 x10-4g / 2.6 g = 3.5 10-4

c = 350 ppm API Interface air intake: as before S/N: 1800:1 LOD: ≈ 600 ppb (single scan)

(M+H)+

background

((H2O)2+H)+

Noise: 1 mV*

Signal: 1.8 V* (M+H)+

background level




Fig. 7 GD-MS accumulated mass spectra of Acetonitril (diluted 1:1000 in MeOH) in ambient air (sample peak at mass 42 not visible here but see fig. 8 below)

Fig. 8 GD-MS accumulated mass spectra of Acetonitril (diluted 1:1000 in MeOH) in ambient air at limit of detection

Analysis Details C.3 cont’d Matrix: ambient air Sample: Acetonitril diluted in Methanol 1:1000 Sample amount: 1 nl in 1 µl Methanol Sample introduction: as before Ionization: as before Sample concentration after injection and evaporation c = 9 x10-7g /2.6 g = 3.5 10-7 c = 350 ppb API Interface air intake: as before

Ion chromatogram of quasi molecular ion of methanol at mass 33 (decline curve)

(M+H)+ MeOH

Analysis Details C.3 cont’d Limit of Detection (LOD) Matrix: ambient air Sample: Acetonitril diluted in Methanol 1:1000 Sample amount: 1 nl in 1 µl Methanol Sample introduction: as before Ionization: as before Sample concentration after injection and evaporation c = 9 x10-7g /2.6 g = 3.5 10-7 c = 350 ppb API Interface air intake: as before LOD: ≈ 100 ppb (accumulated scans

background

(M+H)+ Acetonitril

background

sample




Fig. 9 GD-MS Quasi-molecular ion signal of Acetonitril (diluted 1:1000 in MeOH) in ambient air (SIM mode)

Fig. 10 GD-MS Quasi-molecular ion signal of Acetonitril (diluted 1:1000 in MeOH) in ambient air (SIM mode, cycle time for 2 ions = 5 sec)

Analysis Details C.3 cont’d Selected Ion Monitoring (SIM) Matrix: ambient air Sample: Acetonitril diluted in Methanol 1:1000 Sample amount: 1 nl in 1 µl Methanol Sample introduction: as before Ionization: as before Sample concentration after injection and evaporation c = 9 x10-7g /2.6 g = 3.5 10-7 c = 350 ppb API Interface air intake: as before


(M+H)+ Acetonitril ((H2O)2+H)

Ion chromatogram (M+H)+

10 cycles accumulated

1 cycle = 2.48 sec/ion

(M+H)+ Acetonitril




Fig. 11 GD-MS Quasi-molecular ion signal of Acetonitril (diluted 1:1000 in MeOH) in ambient air (SIM mode) at limit of detection

Fig. 11 GD-MS Quasi-molecular ion chromatogram of Acetonitril (diluted 1:1000 in MeOH) in ambient air (SIM mode) at limit of detection

Signal: 1250 µ V*

Noise: 25 µV*

(M+H)+

S:N = 50:1 LOD: 7 ppb


(M+H)+ Acetonitril

Sample signal

background signal

Analysis Details C.3 cont’d Selected Ion Monitoring (SIM) Limit of Detection (LOD) Matrix: ambient air Sample: Acetonitril diluted in Methanol 1:1000 Sample amount: 1 nl in 1 µl Methanol Sample introduction: as before Ionization: as before Sample concentration after injection and evaporation c = 9 x10-7g /2.6 g = 3.5 10-7 c = 350 ppb API Interface air intake: as before LOD: 7 ppb (SIM mode)




C.3.1 Short-form summary of the applied methodology and the result for GD-MS analysis of Acetonitril (Fig. 4 -12) in ambient air

The compound has been used as an example to demonstrate the measurement procedures for GD-MS trace analysis of a volatile organic compound in ambient air. The rudimentary method used here should give some orientating results for the application of a possible analytical procedure under more sophisticated sample preparation and sample introduction methodology. The counter electrode/nozzle part of the API interface and the API room with ambient air were kept at room temperature. At higher concentrations the volatile compound was injected by a syringe directly into the air of the API room. As described in figures 4 - 6 above an amount of 1 µl Acetonitril was injected and evaporated immediately. The API interface air intake of 0.4 l/min diluted the evaporated sample in the ambient air of the API room (2 l volume) by continuous air exchange from outside. The immediate sample concentration was calculated (see fig. 4) by the ratio of evaporated sample quantity to the air quantity in the API room. According to the decline function about 40% of the sample at higher concentrations may be adsorbed on the inner walls and is desorbing, slowly. A single scan resulted in a Limit of Detection (LOD) of 600 ppb (scan cycle 2.5 sec). At lower concentrations the volatile target compound was diluted 1:1000 in Methanol and 1 µl Methanol (containing 1 nl Acetonitril) was injected by a syringe directly into the air of the 2l API room. In this case a Limit of Detection of 100 ppb (see fig. 7-8) was reached for 24 accumulated spectra (analysis time: 60 sec). In selected ion monitoring (SIM) mode the detection limit (see fig 9-12) was 7 ppb (analysis time: 35 sec). As reported for APCI-MS analyses, previously, it turns out that the individual background signals are the limiting factors (chemical noise) while the sensitivity of the method (number of ions created) would allow even lower limits of detection. Routine applications will require a controlled introduction (gas pumping) of the air matrix containing the analyte gas sample into the counter electrode/nozzle zone of the API interface. The use of external volumes for the sample and the clean flushing gas as well as corresponding switch valves will be mandatory for efficient and precise analyses. All measurements were performed on laboratory ambient air with an approx. water content of 50 %. The results indicate that water acts as a reagent gas which is the basis for the method. The spectra obtained are very similar to those of APCI or classical CI methodologies. The formation of cluster ions of water accompanies the GD-MS method used here. The possibility of using dry air or any other dry matrix gas and exchanging the water by another reagent gas (Methane for instance) for VOC analysis was not investigated within the scope of this application note. Results of some more VOC GD-MS measurements are described in the following figures below yielding similar LOD numbers as for Acetonitril. It will be demonstrated (Isopropanol example) that short analysis times can be reached for one sample by more rapid exchange of the ambient air sample as used in our experiments, usually. This may be of importance for automated routine analysis, if applicable.




D. GD-MS trace analysis of Ethylacetate in ambient air

Fig. 12 GD-MS mass spectrum of Ethylacetate in ambient air (full scan including water cluster formation)

Fig. 12 GD-MS partial mass spectrum with key fragment of Ethylacetate in ambient air

Analysis Details D. Matrix: ambient air Sample: Ethylacetate Sample amount: 1 µl Sample introduction: as before Ionization: as before Sample concentration after injection and evaporation c= 9 x10-4g / 2.6 g = 3.5 10-4

c = 350 ppm API Interface air intake: as before

Ion chromatogram of key fragment at mss 61 (Decline Curve)

(C4H8O2 +H)+

C2H5O2

C2H5O2

(2H2O+H)+

background signal

Analysis Details D. cont’d Limit of Detection (LOD) Matrix: ambient air Sample: Ethylacetate Sample amount: 1 µl Sample introduction: as before Ionization: as before Sample concentration after injection and evaporation c= 9 x10-4g / 2.6 g = 3.5 10-4

c = 350 ppm API Interface air intake: as before Dynamic range : 500:1 LOD estimated: 1 ppm (full scan)




E. GD-MS trace analysis of Acetone in ambient air

Fig. 13 GD-MS accumulated mass spectra of Acetone in ambient air.

F. GD-MS trace analysis of Ethanol in ambient air

Fig. 14 GD-MS accumulated mass spectra of Ethanol in ambient air.

Analysis Details E. Matrix: ambient air Sample: Acetone Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 8 x10-4g/2.6 g = 3 10-4 c = 300 ppm Dynamic range: 1:500 (4 mV peak detectable) LOD: ≈ 600 ppb (accumulated scans)

Ion chromatogram quasi molecular ion of Acetone at mass 59 (Decline Curve)

(M+H)+

accumulated background

Analysis Details E. Matrix: ambient air Sample: Ethanol Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 8 x10-4g/2.6 g = 3 10-4 c = 300 ppm Dynamic range: 1:500 (2 mV peak detectable) LOD: ≈ 600 ppb (accumulated scans)

Ion chromatogram quasi molecular ion of Ethanol at mass 47 (Decline Curve)

(M+H)+





G. GD-MS trace analysis of Isopropanol in ambient air

Fig. 15 GD-MS accumulated mass spectra of Isopropanol in ambient air.

Fig. 16 GD-MS accumulated mass spectra of Isopropanol in ambient air at shortened analysis time

Analysis Details E. Matrix: ambient air Sample: Isopropanol Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 8 x10-4g/2.6 g = 3 10-4 c = 300 ppm Dynamic range: 1:300 (3 mV peak detectable) LOD: ≈ 1 ppm (accumulated scans)

Analysis Details E. Matrix: ambient air Sample: Isopropanol Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 8 x10-4g/2.6 g = 3 10-4 c = 300 ppm LOD: ≈ 1 ppm (same as above)

Ion chromatogram quasi molecular ion of Isopropanaol at mass 61 (decline curve)

(M+H)+


Ion chromatogram quasi molecular ion of Isopropanol at mass 61 (fast decline curve)

(M+H)+




H. GD-MS trace analysis of volatile halogen compounds in ambient air

H.1 GD-MS trace analysis of Dichloromethane in ambient air

Fig. 17 GD-MS accumulated mass spectra of Dichloromethane in ambient air. CH2 is attached to the water clusters

Analysis Details H.2 Matrix: ambient air Sample: Dichloromethane Sample amount: 1 µl Sample introduction: as before Ionization: as before API Interface air intake as before Sample concentration (definition as before) c = 1.3 x10-3g/2.6g = 5 10-4 c = 500 ppm Dynamic range: 1:200 (1 mV peak detectable) LOD: ≈ 2.5 ppm (accumulated scans)


Ion chromatogram of Dichloromethane Isotope at mass 83

molecular ion group ∆m=14


AMD Analysis & Technology Aktiengesellschaft • Königsberger Str. 1 • 27243 Harpstedt Telefon (04244) 968736 • Telefax (04244) 968738 Vorstand: Diplom-Physiker Karl-Heinz Maurer (Vorstandsvorsitzender) Aufsichtsratsvorsitzender: Dipl.-Oec. Mathias Maurer

Application Note 120830 Author: Karl-Heinz Maurer Breath analysis using Atmospheric Pressure Corona Discharge Ionization- Mass Spectrometry (APCDI-MS)

A) Introduction Breath analysis for clinical diagnosis and therapeutic monitoring is considered to be a challenge and research activities by scientific groups are ongoing, continuously. Specific research efforts regarding the analysis of volatile organic compounds (VOCs) by mass spectrometry (MS) in this field have been made since more than one decade and special technologies as proton transfer reaction mass spectrometry (PTR-MS) and selected ion-flow-tube mass spectrometry (SIFT-MS) play a significant role. We have recently reported about analyses performed with the miniaturized AMD Mini QuAS3AR (www.amd-analysis.com) specifically suitable for the direct analysis of low atomic number elements in ESI-MS mode and for direct analysis of volatile organic compounds in ambient air by the application of Atmospheric Pressure Corona Discharge Ionization (APCDI). We were interested to evaluate the potential capabilities of this system for dedicated applications like breath analysis, too.

B) Summary The described analytical results have been performed with the experimental version of a new bench-top double focusing mass spectrometer. The dedicated and miniaturized API-MS system is based on the original AMD QuAS3AR Technology. The API Interface is suitable for ESI-MS, APCI-MS and GD-MS (www.amd-analysis.com). The interface and the mass analyzer are integrated with a multi-stage turbo pumping system. The API Interface was used without any modifications for the breath analyses in APCDI mode. Single breath exhalations were introduced via a Teflon pipe into the API chamber and real-time analyses by full scan techniques were performed. The laboratory air in the API chamber was replaced to a significant portion by introduction of the breath exhalation gas. The water content in the breath gas has been used as a reagent gas for the production of protonated molecular ions of VOCs under atmospheric pressure. The rudimentary sample introduction was accepted due to the orientating character of the analyses which should give an indication of the analytical potential of the system for possible future applications. Due to the limited scope of the project emphasis has been directed to the analysis of the protonated ion at mass 59 of Acetone, known as one of the major components of the human breath and diabetes indicator. No quantifications in this context have been made but estimations of the detection limits for other possibly existing but not identified components (volatile parameters of diagnostic relevance). Determination of these components would require research efforts beyond the scope of this note. Several reproducible measurements have been performed using probands with a normal blood glucose level. We expect that the current detection limits of about 5 ppb can be extended to the upper ppt level using more sophisticated sample introduction methods. Since full scan methodology is the method of choice, extension of the current evaluation system with an array detector for simultaneous ion detection should allow detection limits in the low ppt level using the applied APCDI method and the application of “fingerprint” algorithms” for disease pattern recognition may be possible.


Analysis & Technology Aktiengesellschaft • Königsberger Str. 1 • 27243 Harpstedt Telefon (04244) 968736 • Telefax (04244) 968738 Vorstand: Diplom-Physiker Karl-Heinz Maurer (Vorstandsvorsitzender) Aufsichtsratsvorsitzender: Dipl.-Oec. Mathias Maurer

.

C) Methodology

Fig. 1 In-side view of the API chamber The AMD Mini QuAS3AR mass analyzer, a miniaturized dedicated version of the AMD QuAS3AR double focusing system, was equipped with an API Interface in APCDI mode (APCDI-MS system). System details of the evaluation model have been reported, previously (www.amd-analysis.com). The mass resolution was set to about R= 60 (10% valley). The universal API chamber useable for ESI, APCI, GD and here for APCDI methods consists of a transparent 2 l box with the required infrastructure for the applications in question. In the context of the application described here the major components in use are indicated in the above figure. The corona discharge needle is mounted in front of the counter electrode orifice and used at about + 4000 V, the counter electrode at + 450 V, the nozzle at 0 V and the analyzer is floating on - 2000 V. The temperature of the vacuum part of the API interface was kept at room temperature for the experiment described in figures 2-7 below. The Teflon breath inlet pipe directly used by the exhaling person is not heated and can be exchanged easily for each experiment.

Corona needle Teflon Pipe for introduction of the exhaled breath gas

Orifice of the counter electrode



D) Results D1) Discussion of a selected measurement out of various experiments

Fig. 2 Full scale mass spectra dominated by the protonated water signal and the formation of corresponding water clusters. Above figure describes the initial situation of a mass spectrometric breath measurement using the APCDI method and the standard API interface of the AMD Mini QuAS3AR system. Three single breath exhalations are sequentially introduced by one proband via a Teflon pipe into the API chamber exchanging the residual ambient air in the 2 l volume to a large extent within about 45 sec by the breath gas. Section a) describes the trend of the breath concentration in the API chamber using the ion chromatogram of the protonated Acetone signal at mass 59. The ion chromatogram indicates that the breath gas is replaced by laboratory air with a quasi exponential function. Section b) shows the mass spectrum (raw data) of the breath gas up to mass 100 yielding the proronated water ion at mass 19 and corresponding water clusters at masses 37,55,73 and 91. The quasi molecular Acetone ion is just visible in this scale. The intensity of the water clusters in the breath spectrum is enhanced compared to those in the mass spectrum c) of the ambient air by increased aerosol content.

Ion chromatogram of the Acetone signal at mass 59

Mass spectrum of breath exhalation gas (H2O+H)+ 19

(2H2O+H)+

37 55 73 91

Mass spectrum of laboratory ambient air

a)

b)

c)

Acetone (C3H6O+H)+

Exhalation time



Fig. 3 Extended scale mass spectra showing intense ions of the breath gas in section b) The figure shows that the most significant peak in the breath spectrum, the quasi molecular ion of Acetone, yields an intensity of about 26 mV. The smallest peaks detectable are shown in a figures 5,6 below. The protonated water ion at mass 19 and the water clusters are marked with . . On the one hand side the production of water clusters may interfere with the relevant breath ions but on the other hand it may assist an unequivocal calibration and component identification. The cluster formation can be controlled by the Nitrogen counter flow and the interface temperature. It is reported in the literature that in the breath of healthy persons the Acetone concentration is around 900 ppb. Therefore, it has been assumed in this case that the Acetone signal achieved here corresponds with this value since the blood glucose level of the proband for this experiment was slightly below 100 mg/dL.

a)

b)

c)


Mass spectrum of breath exhalation gas


(C3H6O+H)+



Fig. 4 Normalized scale mass spectra (100% = protonated water ion intensity at mass 19) of breath gas and ambient air

It may be noted from this figure that the highest peak related to the exhaled breath gas, namely the Acetone signal yields about 3 % of the protonated water ion signal as the highest peak in the spectrum. The protonated water ion signal normalized to 100% has got a total signal height of 1V. The data system can handle a 10V signal as maximum, while the smallest signal above base line can be measured on a 10 µV level. The dynamic range for data acquisition therefore is 1:106. For the mass spectra obtained this capacity wasn’t used, fully.




(C3H6O+H)+



Fig. 5 Normalized and further extended scale mass spectra (100% = protonated water ion intensity at mass 19) of breath gas and ambient air

The intensity scale has been extended such that peaks in the order of 1/100 of the Acetone signal can be identified here at a 9 ppb level in the breath gas. The current detection limit for evaluable peaks is demonstrated in the following figure below. Comparing the spectra of ambient air and breath gas at equal scales it is obvious at a first glance that the total sensitivity has increased for breath gas. But besides this, however, the pattern has changed. This is for instance obvious for a number of peaks at masses 15,31,43,44,47,59,63,77,95 which can be assigned to the breath gas. While the existence of the Acetone peak can be considered to be save (confirmed by direct Acetone vapor sniffing) the unequivocal identification of other peaks would require significant research efforts beyond the scope of this note.




(C3H6O+H)+



Fig. 6 Extraction of a portion of the mass spectrum of the breath gas at LOD level. The intensity scale has been extended so far that the limit of detection (LOD) is reached for the current experimental set-up. A peak height of about 200 µV can be evaluated which results in relative terms to the Acetone signal (0.2 / 26 mV) in a ratio of 7.7 x 10-3 x 900 ppb. Current LOD: 6.9 ppb Potential for improvements is available by the application a of more efficient breath sample introduction. In addition it may be mentioned in this context that the measuring time per peak for the applied scan method and mass resolution conditions is in the order of 15 msec while the cycle time is about 3 sec. As a result the duty cycle is about 5 x 10-3. The AMD Mini QuAS3AR analyzer (straight focal plane) is prepared for the implementation of an array detector offering simultaneous ion recording for a wide mass range. The implementation of this technique will improve the duty cycle at least to a value of 50%. This will be an improvement of a factor of 100. The execution of both potential measures should improve the LOD by more than two orders of magnitude. As a result we expect for an optimized breath analyzer on that basis to achieve an LOD in the low ppt range. Perspective LOD: 10 - 20 ppt

Partial mass spectrum of breath exhalation gas



Fig. 7 Ion chromatograms of relevant signals for breath gas spectra and ambient air This figure demonstrates that some useful information may be derived from ion chromatograms. The ion chromatogram of Acetone in trace a) indicates that the background level from the ambient is close to zero and the measured Acetone level clearly belongs to the breath gas. The ion chromatogram of mass 43 in trace b) indicates that a constant level of this ion belongs to the ambient air and the signal increase belongs most probably to the breath gas. However, based on the current information level available it could also belong (at least partially) to an increased sensitivity gain by the higher water content of the breath gas due to optimized reagent gas reaction. At least the constant ion chromatogram of mass 28 in trace c) indicates clearly, that this ion intensity is not influenced by the breath gas..


Ion chromatogram of the signal at mass 43


b)

a)

c)



D2) Result of a specific experiment

Fig. 8 Abnormal Ion chromatogram and mass spectrum In one of our experiments a surprising result occurred represented by above figure. Another proband (blood glucose about 110mg/dl) produced during the on-line breath exhalation a significant amount of saliva which was automatically introduced into the API chamber via the Teflon pipe. After some thoughts we came to the conclusion that the very intensive peak at mass 61 represents the protonated Urea ion. However, medical or other aspects for the existence (e.g. food related) have not been investigated and clarified in the context of this experiment, finally and unequivocal compound identification has not been performed. Therefore, there is some remaining speculative risk for this conclusion. Nevertheless here are some aspects about the occurrence of this ion: Deviating from the experiments described before, the counter electrode and the front vacuum part of the API interface was hold at a temperature of about 1400C in this experiment. Since Urea has got strong water-binding capacity we assume that the protonated Urea ion signal was produced by contact of the saliva with the heated counter electrode. The urea molecule was transferred into the gas phase and water acted as proton donator. The Urea concentration (ppm range) was such high that even the cluster ion at mass 121 was produced. Above considerations may be supported by the fact that the proton affinity (PA) of Urea is significant higher than the PA of water.

(CO(NH2)2+H)+ Urea

Ion chromatogram of the protonated Urea ion signal at mass 61

2(CO(NH2)2)+H)+ M/z = 61



Fig. 9 Ion chromatograms of protonated water, water and Urea The chromatograms in this figure show the trend of relevant ion formations on the time scale. The time period until 135 sec describes the ambient air section. After sample introduction during the period from 135 to 150 sec water content and water protonation increases. After 152 sec a rapid decrease of water protonation combined with a rapid increase of the non protonated water signal and a strong increase of the protonated Urea signal occurs. Beginning at 158 sec until 180 sec a stabilized vapor pressure and proton transfer to the Urea molecule situation takes place. After 180 sec begins a continuous return to the ambient air concentration in the API chamber.

E) Conclusions The target of the measurements performed with the evaluation model of the miniaturized AMD Mini QuAS3AR and described here was to achieve information on the system potential for applications in the health care segment. The project had orientating character and has been performed under limited resources. The simple way of introducing breath gas for online measurements can be improved with relative small technical investments which should contribute to achieve somewhat lower detection limits as reported, currently.

Ion chromatogram of the protonated Urea ion signal at mass 61

Ion chromatogram of the protonated water ion signal

Ion chromatogram of the water ion signal

M/z = 19

M/z = 18

M/z = 61

Sample introduction

stable state



A break-through in this respect is expected by the introduction of an array detector for extraordinary enhancement of the scan duty cycle. The design of the analyzer is prepared for implementation of this attachment which should boost the sensitivity in full scan mode such that detection limits of VOCs in breath gas in the low ppt level should be possible. The question how far higher mass resolution, basically a parameter for improved specificity, will play a significant role besides intelligent pattern recognition (“fingerprint algorithms”) programs may be decided by the ongoing research efforts of the scientific community in this field. We conclude that the results so far achieved with the AMD Mini QuAS3AR equipped with a universal API interface give reason to assume the future applicability of the system for clinical diagnosis and therapeutic monitoring based on breath analysis.



Application Note 120916 Author: Karl-Heinz Maurer Analysis of volatile food aroma compounds by Atmospheric Pressure Corona Discharge Ionization-Mass Spectrometry (APCDI-MS) A) Introduction Food aroma is considered to be an important issue for commercial food producers and consumers as well. Therefore, in this context analysis techniques for identification or confirmation of existence of relevant compounds in food play a significant role. Chromatographic and mass spectrometric methods (specifically PTR-MS) or combination of both (GC-MS) are involved in this field. Many scientific groups world-wide have made efforts by developing sophisticated sample preparations and optimization of analytical instruments for improvement of specificity and limits of detection. We were interested to evaluate how far mass spectra of volatile food aroma compounds, produced in a simple and efficient way without specific sample preparation/enrichment and chromatographic compound separation techniques, can be used for substance classification by pattern recognition methods. In other words it was our aim to evaluate the capability of our miniaturized API-MS, the AMD Mini QuAS3AR mass spectrometer (www.amd-analysis.com) as an “electronic nose” as the method is usually described. B) Summary The described analytical results have been performed again with the experimental version of a new bench-top double focusing mass spectrometer. The dedicated and miniaturized API-MS system (AMD Mini QuAS3AR.pdf) is suitable for ESI-MS, APCDI-MS and GD-MS technologies (www.amd-analysis.com). The interface and the mass analyzer are integrated with a multi-stage turbo pumping system. The universal API Interface was used without any modifications for the volatile food aroma analyses in APCDI mode. The food samples were simply put into the closed API chamber at 25 0C and real-time head space analyses of the released VOC components by full scan techniques were performed. As reported before the water content in the ambient atmospheric air has been used as a reagent gas for the production of protonated molecular ions of VOCs under atmospheric pressure conditions. Due to the orientating character of the analyses, emphasis was given to a rudimentary but very efficient methodology for evaluation of the system potential for future demanding analytical tasks. Therefore, it has been accepted that outstandingly low detection limits were not within the scope of this note. A small number of food examples have been chosen, exemplarily. The mass spectra of the released aroma compounds contained a majority of corresponding protonated molecular ions which could be assigned to volatile flavor components in accord with reliable publicly available information. More extended evaluations and additional supplementary investigations may result in the conclusion that the information content (“fingerprint”) of the obtained flavor mass spectra may be suitable for substance classification by pattern recognition methods. It has been demonstrated that the applied simple and robust APCDI-MS produces mass spectra of volatile compounds for authenticity determinations or quality changes during food storage processes. On-line control of fermentation processes could be another area of application. The use of an array detector for simultaneous ion detection will reduce the analysis time, significantly and make the AMD Mini QuAS3AR system suitable for high through-put automation.

http://www.amd-analysis.com/�





C) Methodology

Fig. 1 Sample directly deposited in the API chamber for direct aroma release The evaluation model of the API-MS mass analyzer (AMD Mini QuAS3AR.pdf) has already been described in various application notes, previously (www.amd-analysis.com). The experimental difference to the ESI, APCI, GD and APCDI applications before is indicated by above picture. The sample for all experiments described here, in this case a Banana, is directly deposited in the API room of 2 l volume at atmospheric pressure and a temperature of about 250C. The aroma compounds are released into the API room (head space) and the corona discharge is initiated for production of protonated molecular ions of the aroma compounds. The corona parameters have been described application notes before. The mass spectra have been recorded in full scan mode and raw data accumulation. The method is very simple and effective.

Corona needle

Orifice of the counter electrode

The Banana deposited in the API room is shown in this picture beside. The Banana is in a full-ripe status, recommended for immediate consumption. Therefore, release of typical banana aroma compounds is expected. The analytical results, obtained within a few seconds or minutes depending on the intensity of the aroma compounds, are described below

Fig. 2




D) Results D1) Banana aroma release

Fig. 3 APCDI mass spectra of full-ripe Banana volatile aroma compounds The upper part of the figure shows the ambient air background spectrum using the same intensity scale as for the banana aroma spectrum. The background spectrum is dominated by the water clusters and the lower intensity trace compounds of the ambient air are not detectable in this scale. The lower part of the figure shows in this scale a number of significant aroma compounds of the banana. It has been proven in previous measurements using the APCDI technique that the VOCs in the gas phase are represented by the M+H ions (proton transfer reaction). The composition of significant aroma compounds in a banana is used from reliable commonly available sources. Therefore, major aroma components above can be identified with a high degree of confidence even if supplementary unequivocal compound identification has not been performed. Some components are summarized in the table below by the M+H ion, compound name and formula.

19

37

55

73

Background mass spectrum

Water clusters corresponding to (H2O+H)+ , (2H2O+H)+ etc.

full-ripe Banana aroma spectrum

Significant aroma compound masses (protonated molecular ions)

159 131 145 117

89

71

61

43

29

159, Isopentyl butanoate (C9H18O2 ); 145, Hexyl acetate (C8H16O2); 131, Isoamyl acetate (C7H14O2 ); 117, Isobutyl acetate, (C6H12O2 ); 89, Ethyl acetate (C4H12O); Other ions occurring at lower masses like 71,61,43,29 have not been assigned to specific compounds.

37 19



Fig. 6 APCDI mass spectra of over-ripe Banana volatile aroma compounds All assigned aroma components of the full-ripe banana are existing in this spectrum. A number of additional ions appear in the higher mass range and it is obvious that low mass components as masses 29, 43, 61 and 89 are more intensive.

Another Banana deposited in the API room is shown in this picture beside. The Banana is in an over-ripe status, probably no more recommended for consumption. Therefore, release of additional compounds supplementary to the typical banana aroma compounds is expected. The analytical results are described below and compared to the first banana.

Fig. 4


Data accumulation time

over-ripe Banana aroma spectrum

145 159 131 117

89

Significant aroma compound masses

71

61 43 29

37

55



Fig. 7 APCDI mass spectra of full-ripe and over-ripe Banana volatile aroma compounds in comparison In this scale the changes in the low mass range (more and more intense peaks for the over-ripe banana) are well visible.





Fig. 8 APCDI mass spectra at extended scale of full-ripe versus over-ripe Banana At extended scale the difference of the aroma compound mass spectra is very well documented. Quantification will require additional investigations regarding careful intensity calibration and normalization to independent relevant ions in the background spectra. Information for creating a so called flavor “fingerprint” and the application of pattern recognition methodologies may be contained in these spectra.





D2) Apple aroma release

Fig. 10 APCDI mass spectrum of the Golden Delicius volatile apple aroma compounds The apple aroma spectrum in the higher mass range shows quite some similarity to the banana spectra. Differences regarding masses and intensities occur mainly in the lower mass range. Some significant ions may be mentioned here as for example 103, 1-Hexanol (C6H14O) and others have not been assigned here.

Golden Delicius apple aroma spectrum 55

37

19

Spectrum at extended scale

159 173

145 131

117

103 85

71 61

57

43 29

The Golden Delicius apple is deposited in the API room as shown in this picture beside. Regarding API infrastructure compare Fig. 1 above. The head space analytical results of the released aroma compounds, are described below.

Fig. 9



Fig. 12 APCDI mass spectrum of the Jonogored apple volatile aroma compounds At a first glance there is no much difference visible in this aroma spectrum compared to the Golden delicious. Therefore, the spectra of both apples components are shown below at extended intensity scale.

The Jonogored apple (shown in this picture beside) is deposited in the API room in the same way as shown in fig. 9 before. The head space analytical results of the aroma compounds, are described below

Fig. 11

37

55

19

Jonogored apple aroma spectrum




Fig. 13 APCDI volatile aroma compound spectra at extended scale of Golden Delicius versus Jonagred apples

Some differences regarding the pattern of the spectra are readily identifiable as indicated above. Quite a number of low intensity peaks are also concerned. However, a reliable difference could possibly be established only by the application of a sophisticated pattern recognition program. The protonated water signal and the protonated water clusters have to be eliminated, naturally.

149

101

99

Jonogored apple aroma spectrum

Golden Delicius apple aroma spectrum



D3) Coffee aroma release

Fig. 15 APCDI volatile aroma compound spectra of roast and ground coffee According to data available, generally there are several hundred volatile compounds in roasted coffee but probably 5% of those may form the so called coffee aroma. In the head space spectrum are a number of significant compounds marked and many other low intensity compounds are existing. Some compounds have been assigned below but a few only may play a role regarding the formation of the coffee aroma. However, the spectrum pattern may be specific for the sample.

A dry portion of roast and ground coffee in a coffee cup is deposited in the API room as shown in this picture beside. Regarding API infrastructure compare Fig. 1 above. The head space analytical results of the released aroma compounds, are described below

Fig. 14

roast and ground coffee aroma spectrum


145

137 131

123 117

109

111

95

97 87

80 81

75 61

57

55

37 19

43

45



Some assigned compounds described by: mass, compound name (formula) 137, Limonene (C10H16)/ 2-Ethyl-3,5-dimethylpyrazine (C8H12N2); 123, 2-Ethyl-5-methylpyrazine (C7H10N2); 97, Furfural (C5H4O2); 87, 2-methylbutanal (C5H10O)/ Diacetyl (C4H6O2); 81, Pyrazine (C4H4N2) 75, Isobutylalkohol (C4H10O); 45, Acetaldehyd (C2H4O);

Fig. 16 APCDI volatile aroma compound spectra of hot coffee vapor

The hot coffee vapor mass spectrum is quite similar to the dry roast and ground coffee spectrum. There are some differences which are described in figure 18 below. It may be remarkable that obviously the much stronger water content of the head space does not influence the spectrum pattern, drastically.

Hot coffee vapor aroma spectrum


19 37

55



Fig. 17 APCDI volatile aroma compound spectra of grain coffee

As expected by a weak smell compared to the roasted coffee beans the intensities of the grain coffee aroma compounds in the head space spectrum are quite low. The background plays a significant role for possible pattern recognition purposes and will complicate the evaluation. A more detailed comparison of the different coffee spectra is shown in figure 18 below.

Grain coffee aroma spectrum

Background spectrum 73 55 37



Fig. 18 APCDI volatile aroma compound spectra at extended scale of roast and ground coffee, hot coffee vapor and grain coffee in comparison

roast and ground coffee aroma spectrum

Grain coffee aroma spectrum

Hot coffee vapor aroma spectrum



D4) Cheese aroma release

Fig. 20 APCDI volatile aroma compound spectra of Young Gouda cheese According to data available, generally, there are around one hundred identified volatile compounds in Gouda cheese. In the head space spectrum are a number of significant compounds marked and many other low intensity compounds are existing which may contribute to the smell of the cheese.

Young Gouda cheese aroma spectrum


Portions of Young and Old Gouda cheese are deposited in the API room, successively. An example is shown in this picture beside. Regarding API infrastructure compare Fig. 1 above. The head space analytical results of the released aroma compounds, are described below

Fig. 19

89

135 159 125 143

107

101 83 77

71 65 59

55 37 19

47

43



Only a few compounds have been assigned below since our interest was not the identification but the pattern of the sample: 107,Benzaldehyde (C7H6O); 89, 3-Methyl-1-butanol (C5H12O); 143, 2-Nonanone (C9H18O);

Fig. 21 APCDI volatile aroma compound spectra of Old Gouda cheese

There are some obvious differences regarding the spectrum pattern between the young and the old Gouda cheese which are better noticeable in the figure 22 below.

Old Gouda cheese aroma spectrum


55 37 19



Fig.22 APCDI volatile aroma compound spectra of Young Gouda cheese versus Old Gouda cheese





D5) Considerations regarding reproducibility and sensitivity of the APCDI method

Fig.23 APCDI low mass section spectra of Young Gouda cheese and Old Gouda cheese in comparison

Major influence on reproducibility and sensitivity of the APCDI method is given by the corona discharge conditions and the water concentration in the gas phase to be analyzed. Since the protonated water acts as a proton donor to the VOC compounds producing (M+H)+ ions in this process, the mechanism requires attention regarding the spectra evaluation. The control of the water cluster formation seems to be important for a reproducible production of the protonated aroma relevant component ions. Above figure shows that the water cluster at mass 55 (green) has been kept at the same intensity for the different measurement of the cheese samples. The intensities of ions (red) belonging to the aroma compounds reflect the differences in reality, at least semi-quantitatively.

Low mass section of Young Gouda cheese aroma spectrum

55 37

19

(2H2O+H)+

(H2O+H)

(3H2O+H)+

47

29

43

55

47

43

29



Fig. 24 APCDI low mass section spectra of Young Gouda and Old Gouda cheese in comparison The corona discharge conditions mainly depend on the position of the corona needle relative to the counter electrode and the applied voltage. Shown in this figure (same spectra as in figure 23 but smaller mass range) two ions seem to be of importance regarding reproducibility of the corona conditions. In all measurements it was observed that the ions at mass 16 and 32 (oxygen), produces in the corona discharge, are not influenced by proton transfer reactions. They may be used as markers for the corona conditions and give at least indications for sensitivity comparisons.

Low mass section of Young Gouda cheese aroma spectrum

Low mass section of Old Gouda cheese aroma spectrum

32 16

16 32



Fig.25 APCDI low mass section spectra of background and hot coffee in comparison

In this figure the background spectrum and the aroma spectrum taken indicate that the corona condition markers show relative deviations in the order of about 20%. Above considerations give some indication that mass spectra produced by controlled APCDI conditions have the potential to be used for pattern recognition purposes.

Low mass section of hot coffee vapor aroma spectrum

Low mass section of background spectrum

32

32 16

16



D6) Considerations regarding applicability of pattern recognition methodologies The application of pattern recognition programs in mass spectrometry is well established since a long time for library search methods based on electron impact (EI) mass spectra yielding significant substance relevant information. Also classification of PTR mass spectra, containing protonated molecular ions of VOC’s is subject of research activities in the scientific community in various fields of application. The target for above described examples of APCDI-MS was the evaluation of the protonated ion spectra regarding the information content (“fingerprint”), possibly suitable for classification of a sample. The bar graph spectra below are shown, exemplarily,

Fig.26 APCDI bar graph protonated ion mass spectra of Young Gouda versus Old Gouda cheese

1.000

10.000

100.000

1.000.000

10.000.000

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

1.000

10.000

100.000

1.000.000

10.000.000

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150





Fig.27 APCDI bar graph protonated ion mass spectra of Young Gouda (blue) versus Old Gouda cheese (red). The raw data spectra of figure 22 have been evaluated at a certain threshold and displayed at logarithmic intensity scale for comparison. Very small peaks are not included and the water clusters at masses 19, 37 and 55 have been eliminated. It can’t be decided from this exemplary figure if the individual “fingerprint” of the different spectra would result in a reliable classification if a pattern recognition program would be applied and the data base for testing the reproducibility would be available. Within the scope of this application note the principal potential of the APCDI-MS method using the headspace of a sample for analysis of the released VOC compounds has been demonstrated.

1.000

10.000

100.000

1.000.000

10.000.000

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

amd analysis & technology ag

Documents