Supplementary Material (SI)

Supplementary material (SI)

for

Optimization of the Ion Source-Mass Spectrometry Parameters in Non-Steroidal Anti-Inflammatory and Analgesic Pharmaceuticals Analysis by a Design of Experiments Approach

Paula Paíga1, Luís M. S. Silva2,* and Cristina Delerue-Matos1

1REQUIMTE/LAQV, Instituto Superior de Engenharia do Porto, Politécnico do Porto, Rua Dr. António Bernardino de Almeida, 431, 4200-072 Porto, Portugal,

2CIETI, Instituto Superior de Engenharia, Politécnico do Porto, Porto, Portugal

Number of pages: 25

Number of tables: 9

Number of figures: 3

Section / Page
Experimental section / SM-3
Plackett-Burman design / SM-3
Complete factorial design / SM-3
Steepest ascent / SM-3
Central composite design / SM-4
Stepwise multiple regression / SM-4
References / SM-5
Tables / Page
Table S1. Chemical structures, physicochemical properties, MS transitions and observed ion ratio for the selected pharmaceuticals for the selected pharmaceuticals / SM-6 a SM-10
Table S2. A 12-run Plackett–Burman design matrix used to identify the main factors of the MS ion source parameters. / SM-11
Table S3. Values at each level for each factor used in Plackett-Burman designs (PB1, PB2, and PB3) and the respectively parameter variation range. / SM-12
Table S4. Ratios of the areas of the studied Plackett-Burman design and the OFAT experiment for each test. The ratios lower than 1, were highlighted with grey color. The concentration of each pharmaceutical was 1 mg L-1. / SM-13
Table S5. Results obtained for the screening of the important factors using the Plackett-Burman design. Data shown for acetaminophen 1 mg L-1. / SM-14
Table S6. First- and second-order functions obtained for each pharmaceutical. / SM-15
Table S7. Analysis of variance (ANOVA) of the first order models with interaction term for the pharmaceuticals tested. / SM-16 a SM-17
Table S8. The experiments used in the face centered central composite design. / SM-18
Table S9. Number of cluster, step number, cluster, and clusters joined for each pharmaceutical based on Ward’s method. / SM-19
Figure / Page
Figure S1. Scheme of the conditions, the experiments performed and the ratio of maximum signals obtained with OFAT and Plackett-Burman design (PB3) approaches. The concentration of each pharmaceutical was 1 mg L-1. / SM-20
Figure S2. Standardized main effects Pareto chart for the Plackett–Burman design. The vertical line in the chart defines the 95% confidence interval. / SM-21 a SM-23
Figure S3. Radar chart with areas and RSD obtained in each run using face centered central composite designs and the sum of the areas for the thirteen pharmaceuticals. / SM-24 a SM-25

Experimental section

Plackett-Burman design

The first step in the optimization strategy was to identify the variables that have significant effect on the response. For screening purpose, the ion source parameters have been evaluated using the Plackett-Burman design (Figure 1a). In this design the interactions are considered to be nonexistent or are negligible relative to main effects [SM1, SM2]. The number of factors and experimental runs are denoted as k and n, respectively, and n must be a multiple of 4 (n=4, 8, 12, 16, …) and has to be greater than k (n>k) [SM2, SM3]. The first task in Plackett-Burman experimental design is to define the k factors to be studied; nonetheless, few test runs as possible are generally desired. Any Plackett-Burman design of n runs can include up to n-1 factors. However, if one includes n-1 factors, there are no degrees of freedom left to estimate the random error variance. With no variance estimate, it cannot test whether any of the factors are statistically significant. So, a careful plan of n runs has to be considered.

Complete factorial design

After the significant factors have been identified, it is necessary to explore the region around the present operating conditions to decide what direction needs to be taken to move towards the optimum region [SM4]. Thus, a projection of the Plackett-Burman design to a full factorial design (Figure 1b) was carried out for the significant variables. This design allowed confirming the significant factors as well as its potential interactions. After the analysis of variance (ANOVA) of the projected factorial design with center runs, one of two conclusions can be drawn concerning the existence of the coefficients affecting pure quadratic terms: (i) there is evidence of rejection of curvature since p-value is higher than 0.05 (Figure 1c) or (ii) there is evidence of curvature because p-value is lower than 0.05 (Figure 1d).

Steepest ascent

Steepest ascent (Figure 1e) or descent, depending on the case, should be applied when initial operating conditions are located far from the region where the factors exhibit curvature for the response of interest. Thus, the objective is to keep experimenting along the path of positive gradient (or negative if the purpose is to obtain the minimum) until there is no further improvement in the response. At that point, a new factorial design (Figure 1b) is conducted to determine a new search direction and repeat the process [SM4]. When the analysis of variance for the quadratic terms of the new model shows statistical significance, a more elaborate experimental design should be conducted incorporating curvature terms to approximate the response. After this an optimal response could be attained.

Central composite design

Central composite designs (Figure 1f) are factorial designs with center points, augmented by a group of axial points (also called star points) that estimate quadratic response surfaces. The central composite design could be efficiently used to estimate first- and second-order terms, to fit second-order response surface models, which comprises a two-level full factorial design with axial points at a distance from the center of the design and center points to predict error variance in a total of 2k+2k+nc where nc represents the number of replications at the center.

Stepwise multiple regression

Stepwise Regression is a the step-by-step iterative construction of a regression model that involves automatic selection of independent variables. Stepwise regression can be achieved either by trying out one independent variable at a time and including it in the regression model if it is statistically significant, or by including all potential independent variables in the model and eliminating those that are not statistically significant, or by a combination of both methods. Stepwise regression method is a combination of forward selection and backward elimination [SM5].

References Supplementary material

[SM1] S. Izadyar, S. Fatemi, T. Mousavand, Synthesis and modification of nano-sized TiO2 for photo-degradation process under visible light irradiation; a Placket–Burman experimental design, Mater. Res. Bull. 2013; 48, 3196.

[SM2] L. Stokes, Sequential Experimentation, Screening Designs, Fold-Over Designs. Available from: faculty.smu.edu/slstokes/stat6337/lecture14.ppt, retrieved on 2014-09-11.

[SM3] J. Tyssedal, Plackett–Burman Designs. Encyclopedia of Statistics in Quality and Reliability, John Wiley & Sons, Ltd., 2008.

[SM4] D.C. Montgomery, Design and Analysis of Experiments third edition ed., John Wiley & Sons, New York, 2001.

[SM5] I.M.M. Ghani, S. Ahmad, Stepwise Multiple Regression Method to Forecast Fish Landing, Procedia Soc. Behav. Sci. 2010; 8, 549.

SM-5

Table S1. Chemical structures, physicochemical propertiesa, MS transitions and observed ion ratio for the selected pharmaceuticals.

Chemical structures and physicochemical properties / Optimized UHPLC-ESI-MS/MS conditions / MRM spectrum
Acetaminophen§
Structure /
CAS number / 103-90-2
Molecular formula / CH3CONHC6H4OH
Molecular weight (g/mol) / 151.16
Grade / Analytical standard
Purity / ≥98%
pKaa / 9.46
log Pa / 0.91
/ Precursor ion (m/z) / 150.20
Product ion
(Quantifier) / Q3 / 107.15
Q1 Pre Bias (V) / 14
CE (V) / 18
Q3 Pre Bias (V) / 21
Product ion
(Qualifier) / Q3 / -
Q1 Pre Bias (V) / -
CE (V) / -
Q3 Pre Bias (V) / -
Ion ratio (± SD n = 6) / -
/
p-AminophenolD
Structure /
CAS number / 123-30-8
Molecular formula / H2NC6H4OH
Molecular weight (g/mol) / 109.13
Grade / n.m.
Purity / ≥99% (HPLC)
pKaa / 5.43; 10.40
log Pa / 0.84
/ Precursor ion (m/z) / 151.00
Product ion
(Quantifier) / Q3 / 109.85
Q1 Pre Bias (V) / -16
CE (V) / -10
Q3 Pre Bias (V) / -23
Product ion
(Qualifier) / Q3 / 64.95
Q1 Pre Bias (V) / -16
CE (V) / -32
Q3 Pre Bias (V) / -24
Ion ratio (± SD n = 6) / 3.98 ± 0.109
/
p-Acetamidophenyl β-D-glucuronide*
Structure /
CAS number / 16110-10-4
Molecular formula / C14H17NO8
Molecular weight (g/mol) / 327.29
Grade / n.m.
Purity / ≥98%
pKaa / 3.17; 12.22
log Pa / -1.04
/ Precursor ion (m/z) / 326.30
Product ion
(Quantifier) / Q3 / 112.95
Q1 Pre Bias (V) / 16
CE (V) / 15
Q3 Pre Bias (V) / 24
Product ion
(Qualifier) / Q3 / 149.90
Q1 Pre Bias (V) / 16
CE (V) / 27
Q3 Pre Bias (V) / 16
Ion ratio ± SD (n = 6) / 2.18 ± 0.007)
/
Ibuprofen§
Structure /
CAS number / 15687-27-1
Molecular formula / C13H18O2
Molecular weight (g/mol) / 206.28
Grade / VETRANAL™, analytical standard
Purity / n.m.
pKaa / 4.85
log Pa / 3.84
/ Precursor ion (m/z) / 205.30
Product ion
(Quantifier) / Q3 / 161.20
Q1 Pre Bias (V) / 13
CE (V) / 10
Q3 Pre Bias (V) / 18
Product ion
(Qualifier) / Q3 / -
Q1 Pre Bias (V) / -
CE (V) / -
Q3 Pre Bias (V) / -
Ion ratio (± SD n = 6) / -
/
Hydroxyibuprofen*, §
Structure /
CAS number / 51146-55-5
Molecular formula / C13H18O3
Molecular weight (g/mol) / 222.28
Grade / VETRANAL™, analytical standard
Purity / nm
pKaa / 4.63b
log Pa / 2.37b
/ Precursor ion (m/z) / 221.30
Product ion
(Quantifier) / Q3 / 177.15
Q1 Pre Bias (V) / 11
CE (V) / 8
Q3 Pre Bias (V) / 21
Product ion
(Qualifier) / Q3 / -
Q1 Pre Bias (V) / -
CE (V) / -
Q3 Pre Bias (V) / -
Ion ratio (± SD n = 6) / -
/
Carboxyibuprofen*
Structure /
CAS number / 15935-54-3
Molecular formula / C13H16O4
Molecular weight (g/mol) / 236.26
Grade / VETRANAL™, analytical standard
Purity / n.m.
pKaa / 3.97b
log Pa / 2.78b
/ Precursor ion (m/z) / 235.30
Product ion
(Quantifier) / Q3 / 191.20
Q1 Pre Bias (V) / 12
CE (V) / 8
Q3 Pre Bias (V) / 14
Product ion
(Qualifier) / Q3 / 72.90
Q1 Pre Bias (V) / 12
CE (V) / 17
Q3 Pre Bias (V) / 15
Ion ratio (± SD n = 6) / 1.39 ± 0.02
/
Acetylsalicylic acid
Structure /
CAS number / 50-78-2
Molecular formula / 2-(CH3CO2)C6H4CO2H
Molecular weight (g/mol) / 180.16
Grade / n.m.
Purity / ≥99.0%
pKaa / 3.41
log Pa / 1.24
/ Precursor ion (m/z) / 179.20
Product ion
(Quantifier) / Q3 / 136.85
Q1 Pre Bias (V) / 22
CE (V) / 10
Q3 Pre Bias (V) / 29
Product ion
(Qualifier) / Q3 / 93.15
Q1 Pre Bias (V) / 22
CE (V) / 22
Q3 Pre Bias (V) / 19
Ion ratio (± SD n = 6) / 1.87 ± 0.03
/
Salicylic acidD
Structure /
CAS number / 69-72-7
Molecular formula / 2-(HO)C6H4CO2H
Molecular weight (g/mol) / 138.12
Grade / n.m.
Purity / ≥99.0%
pKaa / 13.23, 2.79
log Pa / 2.26
/ Precursor ion (m/z) / 137.00
Product ion
(Quantifier) / Q3 / 93.05
Q1 Pre Bias (V) / 18
CE (V) / 18
Q3 Pre Bias (V) / 19
Product ion
(Qualifier) / Q3 / 65.05
Q1 Pre Bias (V) / 18
CE (V) / 31
Q3 Pre Bias (V) / 26
Ion ratio (± SD n = 6) / 12.60±0.248
/
Diclofenac
Structure /
CAS number / 15307-79-6
Molecular formula / C14H10Cl2NNaO2
Molecular weight (g/mol) / 318.13
Grade / n.m.
Purity / ≥98.5%
pKaa / 4.00
log Pa / 4.26
/ Precursor ion (m/z) / 294.20
Product ion
(Quantifier) / Q3 / 249.85
Q1 Pre Bias (V) / 14
CE (V) / 12
Q3 Pre Bias (V) / 18
Product ion
(Qualifier) / Q3 / 34.90
Q1 Pre Bias (V) / 14
CE (V) / 24
Q3 Pre Bias (V) / 13
Ion ratio (± SD n = 6) / 13.04 ±0.04
/
Dipyrone
Structure /
CAS number / 68-89-3
Molecular formula / C13H16N3O4SNa
Molecular weight (g/mol) / 333.34
Grade / Analytical standard
Purity / n.m.
pKaa / 4.85
log Pa / -0.82
/ Precursor ion (m/z) / 310.30
Product ion
(Quantifier) / Q3 / 191.05
Q1 Pre Bias (V) / 16
CE (V) / 15
Q3 Pre Bias (V) / 21
Product ion
(Qualifier) / Q3 / 80.05
Q1 Pre Bias (V) / 16
CE (V) / 3
Q3 Pre Bias (V) / 16
Ion ratio (± SD n = 6) / 2.78±0.027
/
Nimesulide
Structure /
CAS number / 51803-78-2
Molecular formula / C13H12N2O5S
Molecular weight (g/mol) / 308.31
Grade / n.m.
Purity / ≥98% (TLC)
pKaa / 6.86
log Pa / 1.79
/ Precursor ion (m/z) / 307.00
Product ion
(Quantifier) / Q3 / 229.10
Q1 Pre Bias (V) / 10
CE (V) / 18
Q3 Pre Bias (V) / 16
Product ion
(Qualifier) / Q3 / 79.10
Q1 Pre Bias (V) / 10
CE (V) / 29
Q3 Pre Bias (V) / 16
Ion ratio (± SD n = 6) / 7.63 ± 0.05
/
Naproxen
Structure /
CAS number / 22204-53-1
Molecular formula / CH3OC10H6CH(CH3)CO2H
Molecular weight (g/mol) / 230.26
Grade / VETRANAL™, analytical standard
Purity / n.m.
pKaa / 4.19
log Pa / 2.99
/ Precursor ion (m/z) / 229.10
Product ion
(Quantifier) / Q3 / 169.90
Q1 Pre Bias (V) / 10
CE (V) / 17
Q3 Pre Bias (V) / 19
Product ion
(Qualifier) / Q3 / 169.05
Q1 Pre Bias (V) / 10
CE (V) / 35
Q3 Pre Bias (V) / 18
Ion ratio (± SD n = 6) / 1.01 ± 0.01
/
Ketoprofen§
Structure /
CAS number / 22071-15-4
Molecular formula / C16H14O3
Molecular weight (g/mol) / 254.28
Grade / n.m.
Purity / ≥98% (TLC)
pKaa / 3.88
log Pa / 3.61
/ Precursor ion (m/z) / 253.30
Product ion
(Quantifier) / Q3 / 208.90
Q1 Pre Bias (V) / 13
CE (V) / 8
Q3 Pre Bias (V) / 23
Product ion
(Qualifier) / Q3 / -
Q1 Pre Bias (V) / -
CE (V) / -
Q3 Pre Bias (V) / -
Ion ratio (± SD n = 6) / -
/

VETRANAL is a trademark of Sigma-Aldrich Co. LLC