Instrument Detection Limit at Ultrashort Dwell Times Demonstrated on the Agilent 6495C Triple Quadrupole LC/MS

Technical Overview

Authors
Charles Nichols, 
Behrooz Zekavat, 
and Patrick Batoon 
Agilent Technologies, Inc.

Abstract
This Technical Overview presents the measurement of the instrument detection 

limit (IDL) at two different conditions of MRM dwell times: 1) with sufficient ion 

sampling time, and 2) with minimum allowed ion sampling time. The workflow 

used an Agilent 1290 Infinity LC and an Agilent 6495C triple quadrupole LC/MS. 

Results show that even with short MRM dwell times, low IDLs were still confidently 

achievable with this experimental setup.

Instrument Detection Limit at 

Ultrashort Dwell Times Demonstrated 

on the Agilent 6495C Triple 

Quadrupole LC/MS

2

Introduction 
To demonstrate suitability for robustness 

and reliability in routine applications, 

instead of “one-shot” metrics such 

as signal-to-noise (S/N), instrument  

sensitivity is primarily characterized 

using the IDL.1 This specification is more 

rigorous and is determined using the 

ion statistics of replicate injections.2 IDL 

is based on the method detection limit 

(MDL), thoroughly defined by the US EPA 

and other governing bodies.3 “The MDL 

is the minimum measured concentration 

of a substance that can be reported 

with 99% confidence that the measured 

concentration is distinguishable from 

method blank results.” A key difference 

between the IDL and MDL is that the IDL 

is separate from the “analysis method” 

of an application but still determines 

the absolute lowest level of analyte 

the instrument can detect without the 

interference of sample matrix. 
The IDL is the minimum level of analyte 

that results in a statistically differentiable 

signal from the instrument's overall 

noise baseline. Instead of using a 

single injection at high concentration 

(such as is the case with a S/N 

measurement), the IDL is determined 

using replicate measurements at 

low levels, typically with a %RSD of 

10 to 20%. The RSD of response output 

(chromatographic peak area) is used in 

the IDL calculation.
A confidence limit is predefined, 

typically at 99% confidence, for which a 

t-statistic is applied as a multiplication 

factor to determine the IDL on-column 

amount. This confidence arises from 

the statistical variation of a series of 

injections around the instrument’s ability 

to reliably produce replicable data. 

The result of the IDL measurement is 

a statistically meaningful instrument 

performance specification, rather 

than the extrapolation of a single 

measurement at a high injection amount.

The IDL is calculated using the equation:

IDL = (t-statistic) × (amount on column) × (%RSD/100)

The calculated IDL states the minimum 

amount injected on column that can be 

unambiguously distinguished from the 

baseline noise of the chromatograph 

with 99% confidence, not attributable to 

random spikes in noise, or the minimum 

point at which the instrument can reliably 

replicate data. 

Effects of ion sampling 

rate on signal variability: 

Why would the IDL change 

with various dwell times?
An important parameter in the 

discussion of instrument sensitivity is 

the “MRM dwell time”, which influences 

the stochastic sampling of the ion beam. 

Generally, the IDL is characterized by a 

considerable amount of time dedicated 

to sampling the ion beam, allowing a 

stable and consistent flux of ions to 

hit the detector. At lowered instrument 

dwell times (typically <5 ms), ion beam 

sampling becomes less precise due to 

insufficient number of of ions involved in 

the measurement.

Number of ions # ion flux × dwell time

Deficiencies in inadequate sampling 

time (that is, reduced MRM dwell times) 

manifest themselves as decreased 

signal stability (increased %RSD), even 

if the analyte is introduced at a constant 

rate and high concentration. Figure 1 

presents this observation, where the 

%RSD of the ion signal increases at 

decreasing dwell times. 
This Application Note characterizes IDL 

performance when the instrument is 

operated at extremely short dwell times 

(1 and 0.5 ms).
The experiments carried out for this 

paper were acquired on and apply to the 

6495C LC/TQ. Although the performance 

metrics may vary between instrument 

type and model, the overall concept may 

be applied to any other LC/MS or GC/MS 

mass spectrometer.

0%

5%

1%

2%

3%

4%

6%

7%

8%

9%

10%

0

1

2

3

4

5

6

7

8

9

10

RSD 

Dwell time (ms)

Figure 1. ESI-L tune mix infusion acquired at various MRM dwell times. Signal was acquired in MRM mode 
using the transition 622.0 & 622.0 with CE = 0 V. 

3

Experimental

Instrumentation and reagents
• 

Agilent 1290 Infinity binary pump 

(G4220A)

• 

Agilent 1290 Infinity autosampler 

(G4226A)

• 

Agilent 1200 Series autosampler 

thermostat (G1330B)

• 

Agilent ZORBAX Eclipse Plus 

RRHD C18, 2.5 × 50 mm, 1.8 µm 

(p/n 959757-902)

• 

Agilent 6495C triple quadrupole 

LC/MS (G6495C)

• 

ESI/APCI positive ion performance 

standard (G1946-85004)

MRM dwell times for IDL 

measurements
Measurements using the primary 

MRM transition of reserpine 

(m/z 609.3 & 195.1) were acquired at 

200, 1, and 0.5 ms dwell time. However, 

to ensure that the same number of 

chromatographic peak data points 

were collected, a dummy transition of 

m/z 610.3 & 196.1 was included so that 

the overall duty cycle of the instrument 

was 201 ms (shown in Table 2). 

Table 1. LC and MS parameters.

LC Parameters

Flow Rate

0.4 mL/min

Solvent A

H

2O w/ 0.1% FA

Solvent B

ACN w/ 0.1% FA

Gradient

Time (minutes)  %B 

0.00 10 

0.20 10 

1.50 100 

2.50 100 

2.51 10

Stop Time

3 minutes

Post Time

0.30 minutes

MS Parameters

AJS parameters

Sheath Gas Temperature

400 ˚C

Sheath Gas Flow

11 L/min

Drying Gas Temperature

325 ˚C

Drying Gas Flow

11 L/min

Capillary Voltage

4,000 V

Nozzle Voltage

0 V

iFunnel parameters

High-Pressure RF

200 Vpp

Low-Pressure RF

110 Vpp

MRM Transition

609.3 & 195.1

Dummy Transition

610.3 & 196.1

Fragmentor

166 V

Collision Energy

42 V

Duty Cycle

201 ms

Dwell Time

200 , 1, or 0.5 ms

Method parameters

Table 2. Primary and dummy MRM dwell times 
used equating to the same instrument duty cycle 
for all experiments.

MRM Dwell

609.3 & 195.1

Dummy MRM Dwell

610.3 & 196.1

Total Cycle 

Time

200.0 ms

1.0 ms

201.0 ms

1.0 ms

200.0 ms

201.0 ms

0.5 ms

200.5 ms

201.0 ms

4

Results and discussion

Demonstration of IDL using sufficient 

ion sampling 
To demonstrate IDL characterizations 

using the ideal ion sampling scenario, an 

acquisition using an MRM dwell time of 

200 ms was used.
In this case, 1 fg reserpine was injected 

on-column and repeated 10 times 

in MRM mode of acquisition. The 

%RSD of the chromatographic peak 

area was determined to be 17.62%. 

For n = 10 injections, the single-tailed 

t-statistic at 99% confidence was 

found to be 2.821. Putting these 

parameters into the equation, the 

IDL at 200 ms dwell time = 0.49 fg.
Given that the ion beam has been 

sufficiently sampled for 200 ms, 

uncertainty in the measurement arises 

due to the physical absence of ions, 

producing a chromatographic peak 

in close proximity to baseline noise. 

Inspecting the overlaid chromatograms 

in Figure 2, it may appear that injecting 

anything less than 0.49 fg on-column 

would not provide enough ion current.
With injections less than the IDL, signal 

produced will become statistically 

confounded with the chromatographic 

baseline, and it cannot be determined 

with ≥99% confidence that the signal is 

distinct from the noise.

Demonstration of IDL at extremely 

short dwell times 
Referring to Figures 3A and 4A, sufficient 

ion current is produced, demonstrating 

that the instrument can detect these 

on-column amounts with adequate 

“visual” S/N. However, due to insufficient 

ion sampling time, uncertainty in the 

measurement arises from the stochastic 

variation of number of ions hitting the 

detector, thus affecting chromatographic 

peak area variability. 

In line with guidance for characterizing 

IDL, the injected on-column amount of 

reserpine was targeted to produce a 

%RSD between 10 and 20%. For 1 ms 

dwell time, 100 fg reserpine was used, 

resulting in IDL

1 ms = 52.4 fg, while for 

0.5 ms dwell time, 250 fg reserpine was 

used, resulting in IDL

0.5 ms = 119.5 fg.

Both cases produced a chromatographic 

peak far above a reasonable S/N 

threshold, however, the variation 

in chromatographic peak area fell 

between ~40 to 50%, which is deemed 

unacceptable for confident quantitation 

(Figures 3B and 4B). 

When examining additional 

measurements at short dwell times 

far below the characterized IDL 

(10 fg at 1 ms; 50 fg at 0.5 ms), 

some injection replicates produced 

sufficient chromatographic S/N 

(Figures 3C and 4C). However, when 

replicated over a series of injections, 

the chromatographic peak variability 

was obviously unacceptable at 

%RSD≈146.07% and %RSD≈99.35%, 

rendering the data unsuitable for reliable 

quantitative analysis.

Figure 2. Replicate (n = 10) injections of 1 fg reserpine on-column using a 200 ms MRM dwell time.

 Injection No.

Area

1

22

2

20

3

28

4

21

5

25

6

24

7

20

8

14

9

23

10

19

Average

21.6

St. dev.

3.8

%RSD

17.62%

IDL = (2.821) × (1 fg) × (0.1762)
IDL = 0.49 fg on-column

×101

4.8

4.9

5

5.1

5.2

5.3

5.4

5.5

5.6

5.7

5.8

5.9

6

6.1

6.2

6.3

Acquisition time (min)

2.2

2.3

2.4

2.5

2.6

Counts

5

Figure 3. Replicate (n = 10) injections of various amounts of reserpine on-column using 1 ms MRM dwell 
times.

Acquisition time (minutes)

×103

1.3

1.4

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.5

Counts

Acquisition time (minutes)

×103

1.3

1.4

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.5

Counts

Acquisition time (minutes)

1.3

1.4

×103

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.5

Counts

100 fg reserpine

50 fg reserpine

10 fg reserpine

A

B

C

Table 3. Replicate (n = 10) injections of 100, 50, and 
10 fg reserpine on-column using 1 ms MRM dwell 
times.

Injection No. Area (100 fg) Area (50 fg) Area (10 fg)

1

1,791

935

13

2

1,650

448

169

3

1,211

516

202

4

1,902

609

18

5

1,387

189

1

6

1,727

626

11

7

1,159

917

111

8

1,657

836

6

9

2,130

531

0

10

1,747

430

0

Average

1,636.1

603.7

53.1

St. dev.

303.7

236.1

77.6

%RSD

18.56%

39.11%

146.07%

IDL = (2.821) × (100 fg) × (18.56%/100) = 52.4 fg on-column

1 ms MRM dwell time

Figure 4. Replicate (n = 10) injections of various amounts of reserpine on-column using 0.5 ms MRM dwell 
times.

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

1.3 1.35 1.4 1.45 1.5

Acquisition time (minutes)

×103

Counts

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

1.3 1.35 1.4 1.45 1.5

Acquisition time (minutes)

×103

Counts

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

1.3 1.35 1.4 1.45 1.5

Acquisition time (minutes)

×103

Counts

250 fg reserpine

100 fg reserpine

50 fg reserpine

A

B

C

Table 4. Replicate (n = 10) injections of 250, 100, 
and 50 fg reserpine on-column using 0.5 ms MRM 
dwell times.

Injection No. Area (250 fg) Area (100 fg) Area (50 fg)

1

4,868

1,722

511

2

3,118

1,548

8

3

2,825

665

227

4

4,476

477

139

5

4,041

1,292

13

6

4,002

2,072

126

7

3,380

709

82

8

3,769

966

135

9

4,234

2,105

239

10

4,643

2,251

31

Average

3,935.6

1,380.7

151.1

St. dev.

667.1

655.3

150.1

%RSD

16.95%

47.46%

99.35%

IDL = (2.821) × (250 fg) × (16.95%/100) = 119.5 fg on-column

0.5 ms dwell time

www.agilent.com/chem

For Research Use Only. Not for use in diagnostic procedures.

This information is subject to change without notice.

© Agilent Technologies, Inc. 2020 
Printed in the USA, January 29, 2020 
5994-1368EN

Conclusion
This Technical Overview summarizes 

the characterization of the IDL at two 

different MRM dwell time regimes: 

(1) at 200 ms providing sufficient ion 

sampling time; or (2) at 1 and 0.5 ms 

with inadequate sampling time of the ion 

beam. Table 5 shows the results. 
When operated at extremely short 

dwell times such as 1 or 0.5 ms, 

IDL characterization relies on the 

instrument’s ability to reproducibly 

sample and stabilize the ion beam within 

the restricted time frame. Although 

the MRM dwell times are reduced, 

components along the ion optics rail 

were designed to transmit as many ions 

as possible. This results in reasonable 

performance (low IDLs), even at the 

shortest MRM dwell times.

IDL characterization was carried out to 

demonstrate the changes in sensitivity 

with extremely short dwell times 

(demonstrated on the 6495C LC/TQ). 

This was done to provide the customer 

with an expectation of sensitivity 

changes when running extremely 

challenging methods for high-throughput 

and routine applications. 
Although using extremely short dwell 

times is not a generally recommended 

practice, Agilent recognizes that 

customers are facing various challenges, 

demanding scientific accuracy and 

confident results while maximizing 

sample throughput to keep up with the 

cost of running the lab.

References
1.  Sheehan, T.; Yost, R. What’s the 

Most Meaningful Standard for Mass 

Spectrometry: Instrument Detection 

Limit or Signal-to-Noise Ratio? 

Spectroscopy 2018, 13, 16–22.

2.  Wells, G.; et al. Signal, Noise, 

and Detection Limits in Mass 

Spectrometry. Agilent Technologies 

Application Note, publication number 

5990-7651EN, 2011.

3.  Definition and Procedure for the 

Determination of the Method 

Detection Limit, Revision 2. 

Environmental Protection Agency. 

(2016). doi:EPA 821-R-16-006.

Table 5. Replicate (n = 10) injections of 100, 50, and 
10 fg reserpine on-column.

Dwell Time

Injection 

Amount

%RSD

IDL

200 ms

1 fg

17.62%

0.49 fg

1 ms

100 fg

18.56%

52.4 fg

0.5 ms

250 fg

16.95%

119.5 fg