Lauren Brown1; Clara Feider2; Ian Mesher1; Alasdair Edge1; Livia S. Eberlin2
Ambient ionization mass spectrometry has been increasingly used to investigate the distribution of molecules in samples or objects, in situ, without the need for sample preparation. Disadvantages include:• Complex biological samples have inherent matrix e�ects.• No pre-separation means analysis cannot be selective.• Highly abundant analytes may suppress the detection of lower
abundance species.• In tandem mass spectrometry, isobaric interferences can result in
overlapping fragments which present a significant challenge for spectral interpretation.
In-source separation techniques such as field asymmetric ion mobility spectrometry (FAIMS) can be used when chromatographic separation techniques cannot be deployed.• FAIMS separates ions according to field-induced changes in their
collision cross section (Figure 1).
Hypothesized that this is due to the ability to change capillary wall thickness. • Tubing wall thickness of 0.013" was used; thinner than the original
design.• Counteracts potential increased flow from narrower ID.DI2 also allowed the capillary length to be varied with ease.• The MRFA signal intensity between DESI, DI1 and two DI2
capillaries (5” and 2”) was compared (Figure 11).
The longer DESI-FAIMS capillaries showed little transmission of the doubly charged m/z 262 ion, but the signal was more abundant with the shorter capillary. • Possibly due to greater path travelled by the ions resulting in
charge reduction. • Suggests capillary could be tailored to the application.
• As the field reverses polarity and changes magnitude, the ion changes direction and speed.
• Each ion has specific net sideways drift velocity.• By applying a DC compensation field (CF), this drift can be
corrected, focussing ions through the device.• Analogous to atmospheric pressure quadrupole.• Highly orthogonal to m/z separation o�ering a means to increase
peak capacity for the analysis of complex samples by ambient ionization mass spectrometry.
• Can selectively transmit ion of interest prior to mass spectral detection.
A recent study1 demonstrated the use of FAIMS in combination with desorption electrospray ionization (DESI) (Figure 2).
• Inner diameter (ID) of 0.067” was used to best replicate the standard FAIMS inlet ID.
• Previously optimised for flow rate through the FAIMS chip for optimal full width at half maximum (FWHM)2.
• Di�erent thickness walls were tested (Figure 6).
• Having the capillary attached to a Swagelok nut allows it to be rotated upon its axis.
• Allows adjusting of capillary angle towards stage.
• When nut is tightened it will lock its orientation.
• Thinnest walls gave best sensitivity• Direction is fixed as a result of being welded• Small changes in capillary direction could account for di�erences
in signal intensity.Options:1. Fix the capillary in the optimal location.2. Make the direction of the bend adjustable (Figure 7).
A modified FAIMS inlet (DI2) was designed to take a 1/16” Swagelok nut (Figure 8).
• DESI-FAIMS-MS resulted in enhanced mass spectral detection compared with DESI-MS alone.
• Increased s/n due to suppressed background.• Increased absolute signal as trap was preferentially filled with
FAIMS-transmitted low abundance ions, rather than naturally high abundance species.
Here we describe optimisation of the mechanical integration of a desorption electrospray ionization (DESI) system, with FAIMS.
Key goals of the design were to:• Deliver a carrier gas velocity profile through the FAIMS chip
optimized for high separation performance. • Minimize transmission losses.In the first design iteration (DI1), an unheated extended transfer tube of 5” was welded on to the FAIMS inlet (Figure 5).
1. Introduction
CF sweeps were acquired by rastering across a mouse kidney section, selected due to its high homogeneity, with both the DI1 and DI2 capillaries.Increased resolution between the doubly and singly charged cardiolipin peaks was achieved using the DI2 capillary due to narrower, more Gaussian FAIMS peaks (Figure 10).
4. DESI-FAIMS data
ultraFAIMS provides a practical means of adding additional in-source separation for ambient ionisation mass spectrometry.The optimised DI2 inlet resulted in increased resolution and allowed sensitivity to be optimised via use of different length capillaries and wall thicknesses. Future experiments will look at increasing performance further by investigating different dimension inlets and capillaries, and different capillary lengths and materials.
5. Conclusions
To integrate the ultraFAIMS device and Omni Spray source an extension to the transfer capillary is required (Figure 4).
It was expected that the two di�erent DESI-FAIMS interfaces may exhibit di�erences in flow through the FAIMS chip.• 1/16” OD capillary will have slightly smaller ID compared to front
insert (Figure 9).
• The narrower ID may produce higher transmission and wider FAIMS peak FWHM, due to an increase in flow rate through FAIMS device.
3. DESI-FAIMS interface designs
Figure 1. FAIMS separates ions according to field-induced changes in their collision cross section
Figure 2. Schematic representation of selective FAIMS transmission of ion of interest using DESI-FAIMS-MS
Figure 3. (a) ultraFAIMS T1 device and (b) installed on Q-Exactive with 2D Omni Spray
www.owlstonemedical.com/ultraFAIMS
Additional Separation for Ambient Ionisation Mass Spectrometry:The Development of a DESI-FAIMS-MS Interface
References1. Feider, C.L.; Elizondo, N.; Eberlin, L.S.; Anal. Chem. 2016 88 (23)2. Brown et al.; Optimization and performance characterization of a
microscale FAIMS chip coupled to an Orbitrap mass spectrometer. Poster presented at 62nd ASMS Conference on Mass Spectrometry and Allied Topics, 2014, June 15 - 19, Baltimore, MD
1Owlstone Medical Ltd., 162 Cambridge Science Park, Cambridge, CB4 OGH, UK, 2University of Texas at Austin, Austin TX, 78712, USAFor further information, email: [email protected]
Experiments were carried out using an ultraFAIMS-T1 system (Owlstone Medical Ltd., Cambridge, UK) installed on Q-Exactive and Thermo Elite mass spectrometers (Thermo Scientific, San Jose, CA) with a 2D Omni Spray (Prosolia Inc., Indianapolis, IN) (Figure 3).ultraFAIMS is a miniaturised FAIMS device in which the electrodes are formed from a micro-manufactured chip.
2. Methods
ION COLLISION CROSS SECTION VARIES AT HIGH VSLOW DISPERSION FIELDS, GIVING DIFFERENT MOBILITIES
CompensationField 1
CompensationField 2
CompensationField 3
DispersionField (DF)
Electrode
Electrode
Electrode
Electrode
Carrier Gas
Electrode
Electrode
0 Time
Low field
High field
Ion
Ion
Ions enterchannel
todetector
Ions enterchannel
todetector
Ions enterchannel
todetector
30.0° ±0.05
Ø 1
.6R
EA
M
0.2 45.0°
Capillary extention weldedon to FAIMS inlet
X 2 POS
Ø 4
.3 ±
0.0
5
M8 x 1.0
Figure 5. (a) ultraFAIMS T1 device and (b) installed on Q-Exactive with 2D Omni Spray
Figure 4. Standard FAIMS inlet design indicating part needing extension
Figure 6. Comparison of thickest and thinnest walled capillaries with FAIMS voltages set to transmit cardiolipins
Figure 7. Rotation of capillary towards stage via Swagelok nut
Figure 8. 1/16” Swagelok nut DESI-FAIMS interface design (DI2)
0.020”
650 700 750 800 850 900 950 1000m/z
0
20
40
60
80
100
0
20
40
60
80
100
Rel
ativ
e A
bun
dan
ce(%
)
885.550
834.530
749.494
738.502761.494
862.653724.486 790.540
891.643
713.494
810.529
917.478
749.494
885.550737.494 834.530
761.494
724.486
773.493 844.644806.546
891.642
713.494917.985
m/z749.4
95
m/z738.5
02
m/z761.4
94
m/z749.4
95
m/z738.5
02
m/z761.494
Thin (0.011 in.) NL=1.57E 4
Thick (0.2 in.)
NL=4.64E 3
0% Relative
Abundance 3 mm
100%
(DF=220 Td, CF=2.20 Td)
1/16” OD Capillary1/16” Swagelok nut
1/16” Swage cone
Screws onto ourpiece – no solderrequired
1/16” OD Capillary willhave slightly smaller IDcompared to our02-1519 part
1.6mm ID THRU
M8 x 1.0
Ø
3.0
(52
.0°)
Ø 4
.3 ±
0.0
5
30.0° ±0.05°
Ø 1
.6 T
HR
U 0.2 X 45°
3 POS
0.4 Ra
Figure 9. Customised 1/16” Swagelok nut DESI-FAIMS inlet design
Figure 11. MRFA signal intensity between DESI, DI1 and 5” and 2” capillary DI2 inlets
Figure 10. DESI-FAIMS spectra of doubly and singly charged cardiolipin peaks from DI1 (left) and DI2 (right).
CF Sweep Start
CF Sweep End
RT:0.00-4.27
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2
Time (min)
0
10
20
30
40
50
60
70
80
90
1000
10
20
30
40
50
60
70
80
90
100
Rel
ativ
e A
bund
ance
NL: 2.25E2m/z= 1449.57-1450.39 MS 07252017_oldFAIMS_Kidney1_DF210CFneg1to3_240sec_IT
NL: 1.31E3724.17-724.65 MS 07252017_oldFAIMS_Kidney1_DF210CFneg1to3_240se
CF=1.38 Td
CF=1.71 Td
Doubly Charged
CL
Singly Charged
CL
CF Sweep Start
CF Sweep End
CF=1.06 TdRT:0.00-4.15
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0
Time (min)
0
10
20
30
40
50
60
70
80
90
1000
10
20
30
40
50
60
70
80
90
100
Rel
ativ
e A
bund
ance
NL: 3.70E21449.52-1450.33 S_logcap_Kidney1_40sec_IT
NL: 7.88E2MS S_logcap_Kidney1_40sec_IT
CF=1.94 Td
Doubly Charged
CL
Singly Charged
CL
All data 210 Td DF, -1 to + Td CF
200 400 600 800 1000 1200 1400 1600 1800 20000
10
20
30
40
50
60
70
80
90
100
Rel
ativ
e A
bun
dan
ce 524.3333
262.7500Regular DESI CapillaryNL=4.94E6
200 400 600 800 1000 1200 1400 1600 1800 2000
m/z
0
10
20
30
40
50
60
70
80
90
100
Rel
ativ
e A
bun
dan
ce
262.7500
524.3333Old DESI-FAIMS CapillaryNL=3.59E5
200 400 600 800 1000 1200 1400 1600 1800 2000
m/z
0
10
20
30
40
50
60
70
80
90
100
Rel
ativ
e A
bun
dan
ce
524.3333
262.7500
New DESI-FAIMS Capillary (5in)NL=4.24E5
m/z
200 400 600 800 1000 1200 1400 1600 1800 2000m/z
0
10
20
30
40
50
60
70
80
90
100
Rel
ativ
e A
bun
dan
ce
524.4167
262.7500
New DESI-FAIMS Capillary (2in)NL=8.64E5
(a) (b)
N2
MovingStage
X
Y
Extended Capillary
FAIMS Coneand Chip Mass Spectrometer
m/z
DesorbedIons
VSolvent
