This work demonstrates my understanding of matrix effects.
A set of experiments was designed to investigate three internal standards. It is the objective to select one of them to correct for losses during sample processing and to mitigate the impact of matrix effects on quantitation by mass spectrometry. The data were provided as follows:
Three strategies for using internal standards were studied. One internal standard is used for one strategy; Two internal standards are considered for two strategies. Further work is taken into consideration.
A couple of conclusion could be made according to the retention times in table I. The internal standard option 1 (option 1) would be chosen because its retention time is the closest to the analyte, so its polarity would be similar to the analyte, and it may be a structure analog of the analyte. Internal standard option 2 (option 2) would be more polar than the analyte, assuming that a reverse-phase C18 column is used.
Compounds |
Retention Time (min.) |
---|---|
Analyte |
5.43 |
Option 1 |
5.82 |
Option 2 |
2.36 |
Option 3 |
4.34 |
Matrix effects were evaluated by comparing the peak areas for the analyte and three internal standards options in post-extraction and solvent. The ratio for the analyte is 1.10, indicating ion enhancement, and the ratio for three internal standard options ranged from 0.22 to 0.98 (Table II). Option 2 had significant ion suppression. Matrix effects for option 1, internal standard option 3 (option 3), and the analyte were similar.
Compounds |
Peak Area in Solvent |
Peak Area in |
\(\frac{Peak Area_{post-Extraction}}{Peak Area_{solvent}}\) |
---|---|---|---|
Analyte |
613,548 |
672.152 |
1.10 |
Option 1 |
901,751 |
879,563 |
0.98 |
Option 2 |
1,154,890 |
256,987 |
0.22 |
Option 3 |
798,521 |
768,243 |
0.96 |
Option 3 had an excellent recovery (102%), which indicated the sample processing had no effect on it. Therefore, it can’t be used to correct the losses. On the contrary, option 2 has a similar recovery with the analyte (Table III). It can be used for correcting the losses due to sample processing.
Compounds |
Peak Area ± RSD% |
Peak Area in |
Recovery % |
---|---|---|---|
Analyte |
556,774.2 ± 4.16% |
672,152 |
82.8%
|
Option 1 |
585,436.6 ± 2.06% |
879,563 |
66.6% |
Option 2 |
218,909.6 ± 3.50% |
256,987 |
85.2%
|
Option 3 |
785,597.4 ± 1.45% |
768,243 |
102.26% |
Three strategies were evaluated.
Option 2 is used to correct the losses and mitigate the matrix effects because its recovery is close to that of the analyte, even though the matrix effect is the greatest.
The method is described as follows:
1) The sample is fortified with option 2 prior to sample
processing.
2) Calibration standards are prepared in matrix blank.
3) Option 2 is added to each calibration standard solution prior to
analysis. The peak area of option 2 in calibration standard is used for
recovery correction.
Firstly, calculate the percentage recovery of option 2.
Peak Area |
Peak Area |
Recovery % |
---|---|---|
221,643 |
256,987 |
86.2 |
222,650 |
256,987 |
86.6 |
226,831 |
256,987 |
88.3 |
216,583 |
256,987 |
84.3 |
206,841 |
256,987 |
80.5 |
The recovery of option 2 ranged from 80% to 88%.
Secondly, calculate the recovery corrected concentration of the analyte.
The analyte peak area relative to option 2 peak area (post-extraction) used to build the calibration curve is: 2.616.
Peak Area |
Conc. % |
Conc. % |
---|---|---|
553,428 |
82.3 |
95.5 |
536,132 |
79.8 |
92.1 |
593,679 |
88.3 |
100.1 |
561,770 |
83.6 |
99.2 |
538,862 |
80.2 |
99.6 |
The mean ± standard deviation percentage corrected recovery is: 97 ± 3%.
There are a couple of reasons to select option 2 as a one-internal
standard strategy for both correcting the losses and mitigating matrix
effects to quantify analysis.
(1) The mean ± SD percentage (97 ± 3%) is acceptable, even though its
significant ion suppression (the peak area of it in the post extraction
relative to the peak area of it in the solvent is 0.22).
(2) Such a strategy has been found in the literature (Tittlemier et
al., 2016).
One concern, though, is that the variability of option 2 would be noticeable if the sample matrix fluctuates in different samples due to its matrix effect. I am interested in its structure. Further work on the variability of option 2 in replicates on different days would be taken into consideration.
Option 2 is used to correct the losses, and option 1 is used to mitigate the matrix effects.
The empirical principle supports the practice of using option 1 as an internal standard. Option 1 would be a structure analoge of the analyte for quantification because it elutes at about the same time (5.82 min.) as the analyte (5.43 min.), and they have similar matrix effects. It can’t be used for correcting the losses in sample processing because its recovery (67%) is lower than that of the analyte (83%).
The benefit of using option 2 for correcting the losses is its recovery is close to that of the analyte. The benefits of using option 1 is: It could be a structure analog of analyte. Their retention time is close, the matrix effects for both of them are similar.
The method is described as follows:
(1) Sample is fortified with option 2 prior to sample processing.
(2) Calibration standards are prepared in matrix blank.
(3) Option 1 is added to the sample as well as each calibration
standard solution prior to analysis.
The result of using the pair of two internal standards turned out the same as the result of using single option 2 (Table VIII).
Interestingly, curiosity took me to exam what is the outcome if the pair of option 2 and option 3 is used. Option 2 is used to correct the losses, and the option 3 is used to mitigate the matrix effects. The benefit of using option 3 is that the peak area of pre-extraction and the peak area of post-extraction are close.
The method is described as follows:
(1) Sample is fortified with option 2 and option 3 prior to sample
processing.
(2) Calibration standards are prepared in matrix blank.
(3) Option 3 is added to each calibration standard solution prior to
analysis.
The peak area of option 2 in post extraction relative to that of option 3 peak area in post-extraction used to build the calibration curve is: 0.3345.
The peak area of analyte in pre-extraction relative to that of option 3 peak area in post-extraction used to build the calibration curve is: 0.8749.
Option 2 Peak Area |
Option 3 Peak Area |
Option 2 |
---|---|---|
221,643 |
789,654 |
83.9% |
222,650 |
776,321 |
85.7% |
226,831 |
791,622 |
85.7% |
216,583 |
771,423 |
83.9% |
206,841 |
798,967 |
77.4% |
Analyte Peak Area |
Option 3 Peak Area |
Conc.% |
Recovery % |
---|---|---|---|
553,428 |
789,654 |
82.3 |
98.1% |
536,132 |
776,321 |
79.8 |
93.0% |
593,679 |
791,622 |
88.3 |
103.1% |
561,770 |
771,423 |
83.6 |
99.6% |
538,862 |
798,967 |
80.2 |
103.6% |
The outcome using this pair of two internal standards gives is: The mean ± SD percentage: 100 ± 4%.
Option 3 gives an ideal result in the strategy of using two internal standards. The mean ± SD percentage (100 ± 4%). It is the best candidate for quantification because its retention time (4.34 min.) is not far away from the analyte’s retention time (5.43 min.); the matrix effects for it and the analyte is similar; and its recovery during the sample processing is the best (102%).
The comparison of the three strategies’ results is summarized in table VIII.
Internal Standard(s) |
Mean ± SD Percentage |
---|---|
Option 2 |
97 ± 3% |
Option 2 for correcting losses |
97 ± 3% |
Option 2 for correcting losses |
100 ± 4% |
The selection of option 2 and option 3 combination gave the highest accurate result with same high precision.
If we want to go with two internal standard strategy, option 2 with option 3 is the answer.
When screening the internal standards for one internal standard strategy, the recovery is always the first criterion.
Created Date: 2023-01-23
Last Modified Date: 2023-01-23