Advertisement

Chromatographia

, Volume 82, Issue 1, pp 221–233 | Cite as

Mass Spectrometry-Compatible Enantiomeric Separations of 100 Pesticides Using Core–Shell Chiral Stationary Phases and Evaluation of Iterative Curve Fitting Models for Overlapping Peaks

  • Garrett Hellinghausen
  • Elizabeth R. Readel
  • M. Farooq Wahab
  • J. T. Lee
  • Diego A. Lopez
  • Choyce A. Weatherly
  • Daniel W. ArmstrongEmail author
Original
  • 223 Downloads
Part of the following topical collections:
  1. 50th Anniversary Commemorative Issue

Abstract

Pesticides are often chiral, and their isomers have different activity, toxicity, metabolism, and degradation properties. Perhaps, the most complex are the synthetic pyrethroid insecticides that have up to 8 stereoisomers, but not all are active. Pyrethroids are toxic to aquatic invertebrates and non-targeted species like honey bees since they persist in the environment. Extensive biological studies of the pyrethroid enantiomers are limited. Possibly, this is because liquid chromatography enantiomeric methods for these studies often have limitations with mass spectrometry (MS) compatibility. In this study, an effective methodology was developed with MS compatible solvents to evaluate several core–shell (superficially porous particle, SPP) chiral stationary phases (CSPs) for the enantiomeric separation of several classes of chiral pesticides. The CSP with the broadest selectivity or spectrum amongst all pesticide classes was the hydroxypropyl-β-cyclodextrin. The other CSPs (cyclofructan, macrocyclic glycopeptide, and quinine-based selectors) had more selective applications including separations of the pesticides with amine or acid functionalities. Overall, 74 of 100 pesticides were baseline-separated. Most of the remaining ones had multiple stereogenic centers and had only one overlapping pair. Such cases were evaluated with a convenient peak area extraction protocol by iterative curve fitting. This approach will lead to more facile enantiomeric analyses where MS is needed to overcome complex matrices and reduce extensive method optimization.

Graphical abstract

Keywords

Fungicides Herbicides Insecticides Pyrethroids Superficially porous particles (SPP) 

Introduction

Pesticides are substances used for controlling, preventing, or destroying animal, microbiological or plant pests [1]. Most commercial pesticides are synthetic and can be categorized by their activity as insecticides, rodenticides, fungicides, herbicides, etc. They are also commonly categorized by their structures, including pyrethroids, organophosphates, acylanilides, triazole-related fungicides, imidazolidinones, and phenoxypropionic acid herbicides. Many are chiral and commercialized as racemic mixtures. However, their isomers have differences in terms of metabolism, toxicity, carcinogenicity, activity, etc. In the 1950s, R enantiomers of phenoxypropionic herbicides like dichlorprop and mecoprop were found to be more active than their respective S antipodes [2]. Their enantiomers also biodegrade differently [3]. One of the four isomers of the fungicide, triadimenol, was found to be 1000-fold more active than the other three [4]. These enantiomeric studies have led to more effective commercial pesticides with selected active isomers, a procedure analogous to a “chiral switch” in the pharmaceutical industry [5]. Also, it limits unnecessary pollution found in foods, beverages, and the environment [6, 7]. While metabolism and risk to the environment have sometimes been assessed for individual pesticide enantiomers, very little is known for those with multiple chiral centers [8, 9, 10].

Perhaps, the most complex chiral pesticides are the pyrethroid insecticides, which commonly have two to eight stereoisomers [11]. Pyrethroids are popular due to their high specificity, activity, low toxicity to mammals, and UV-resistance compared to other insecticides such as organophosphates and natural pyrethrins (from Chrysanthemum cinerariaefolium) [11]. However, they persist, and pollute the environment, specifically crops and small aquatic reservoirs [11]. Most are deadly to aquatic species, especially bottom-feeder scavenger fish and other non-target organisms like honey bees and butterflies [12, 13, 14, 15]. Their enantiomers are likely to have different effects on non-target organisms. In general, the cis-isomers of pyrethroids are more toxic than the trans-isomers [11]. However, trends are not as general for the pyrethroid’s various S and R configurations. For example, 1R-cis-permethrin and 1S-syn-fenvalerate have the highest acute toxicity, compared to their other three respective stereoisomers [12, 13]. These studies might be important to the continual prevention of insecticidal diseases, like malaria [16]. Recently, pyrethroid resistance has increased, which has led to new pyrethroid-based interventions [16]. Therefore, with adequate enantiomeric analysis, new pyrethroid stereochemical properties could be thoroughly understood so the applications could be more effective and environmentally friendly.

Enantiomeric pesticide analysis is commonly performed by LC for high sensitivity and GC, but GC might not be best to study enantiomeric properties since pesticide enantiomers are susceptible to isomerization and epimerization at elevated temperatures [17, 18]. With LC, the most commonly reported enantiomeric pesticide separations utilized π-complex-type, and polysaccharide-based chiral stationary phases (CSPs), while some are performed by cyclodextrin-based CSPs [19, 20, 21, 22, 23, 24, 25, 26]. Some approaches used reversed phase solvents, which facilitated enantiomeric studies of some triazole fungicides with two to four stereoisomers [23, 24]. However, for other pesticides (up to 8 stereoisomers), normal phase solvents/multiple CSPs coupled in tandem usually were required [19, 20, 21, 22]. Recently, improved methodologies utilizing core–shell or superficially porous particle (SPP) CSPs to increase efficiency and shorten analysis times were developed [27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37]. Also, they are compatible with mass spectrometry (MS) detection, which is especially significant when sensitive pesticide analysis is needed, using new MS techniques like paired-ion electrospray ionization (PIESI) [38]. We effectively utilized screening and optimization methodologies for 100 chiral pesticides with 6 SPP CSPs: hydroxypropyl-β-cyclodextrin (CDShell-RSP), isopropyl-cyclofructan 6 (LarihcShell-P), quinine-based CSP (Q-Shell), and three macrocyclic glycopeptides (NicoShell, TeicoShell, and VancoShell) [27, 28, 29, 30, 31, 32, 39, 40, 41].

Since baseline separation of all isomers for complex pesticides with multiple chiral centers can be difficult to achieve, an iterative curve fitting method was used which recovered the overlapping peak areas with high accuracy. Deconvolution techniques are usually utilized to remove band broadening and other distortions but can be used to determine areas of overlapped peaks with errors less than or comparable to other peak integration methods [42, 43, 44, 45]. Commercial software is available for curve fitting, but it can now be performed using ubiquitous Microsoft Excel [46]. Often, these methods are used for separations from complex matrices, as in omics-related fields [45, 46]. More recently, they have been utilized in ultrafast chiral chromatography and highly accurate enantiomeric fraction analysis of pesticides [47, 48, 49]. The protocol utilized in this study is thoroughly described.

Experimental

Chemicals and Materials

(R,S)-Hydroxypropyl-β-cyclodextrin (CDShell-RSP, RSP), teicoplanin (TeicoShell, TS), modified macrocyclic glycopeptide (NicoShell, NS), vancomycin (VancoShell, VS), quinine-based (Q-Shell, QS), isopropyl-cyclofructan 6 (LarihcShell-P, LSP) were obtained as 2.7 µm core–shell (superficially porous particles) 100 × 4.6 mm (i.d.) and 150 × 4.6 mm (i.d.) columns from AZYP, LLC. (Arlington, TX, USA). The packing procedure of the 150 × 4.6 mm (i.d.) CDShell-RSP column was slightly modified, especially for the pyrethroids with multiple stereogenic centers. Analytes were purchased as racemic standards from Sigma-Aldrich (St. Louis, MO, USA), Toronto Research Chemicals (Toronto, CN), and LKT Laboratories Inc (Minneapolis, MN, USA). Racemic standards were prepared with methanol at 1 mg/mL for analysis. Solvents and additives including HPLC grade acetonitrile (ACN), methanol (MeOH), ethanol (EtOH), isopropyl alcohol (IPA), tetrahydrofuran (THF), hexane (Hex), tert-butyl methyl ether (MTBE), acetic acid (HOAc), trifluoroacetic acid (TFA), trimethylamine (TEA), formic acid (FA), and ammonium formate (NH4HCO2) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Water was purified by a Milli-Q water purification system (Millipore, Billerica, MA, USA).

Chromatographic Conditions

An Agilent 1260 (Agilent Technologies, Palo Alto, CA, USA) HPLC was used for all studies. It consisted of a 1200 diode array detector, autosampler, and quaternary pump. The mass spectrometer used in this study was a Shimadzu triple quadrupole liquid chromatography-mass spectrometry (LC–MS) instrument, LCMS-8040, (Shimadzu, Tokyo, Japan). All MS was operated in positive ion mode with an electron spray ionization source. The parameters were set as follows: nebulizer gas flow, 3 L/min; drying gas flow, 15 L/min; desolvation line temperature, 250 °C; heat block temperature, 400 °C. Multiple UV wavelengths, 220, 230, and 254 nm were utilized for detection and identification of enantiomers. Most separations were carried out at 25 °C, unless otherwise noted, using an isocratic method. Mobile phases were degassed by ultrasonication under vacuum for 5 min. Each analyte was screened in polar ionic mode (PIM), polar organic mode (POM), reversed phase (RP), and normal phase (NP) following previous protocols [31, 32, 39, 40, 41]. The optimized conditions as labeled in Table 1 were: PIM1(a,b,c,d): 100:(0.5,0.2,0.1,0.025) MeOH:NH4HCO2 (v/w), PIM2: 100:0.2:0.05, MeOH:HOAc:NH4OH (v/v/v), POM1: 60:40:0.3:0.2, ACN:MeOH:HOAc:TEA (v/v/v/v), RP1(a,b,c,d): (90:10,60:40,40:60,30:70), MeOH:16 mM NH4HCO2 pH 3.6 (v/v), RP1(e): 30:70, MeOH:48 mM NH4HCO2 pH 3.6 (v/v), RP2(a,b,c,d,e,f): (30:70,25:75,20:80,15:85,10:90,5:95), ACN:16 mM NH4HCO2 pH 3.6 (v/v), RP2(g): 20:80, ACN:48 mM NH4HCO2 pH 3.6 (v/v), RP2(h): 20:80:0.005, ACN:16 mM NH4HCO2 pH 3.6: MTBE (v/v/v), RP2(i): 15:85, ACN:48 mM NH4HCO2 pH 3.6 (v/v), RP3(a,b,c): (40:60,35:65,20:80), EtOH:16 mM NH4HCO2 pH 3.6 (v/v), RP4(a,b): (30:70,20:80) IPA:16 mM NH4HCO2 pH 3.6 (v/v), RP5: 10:90, THF:16 mM NH4HCO2 pH 3.6 (v/v), NP1: 95:5, Hex:IPA (v/v), NP2: 90:10, Hep:EtOH (v/v), NP3: 90:10:0.3:0.2, Hep:IPA:TFA:TEA (v/v/v/v), NP4: 70:30:0.3:0.2, Hex:EtOH:TFA:TEA (v/v/v/v).

Table 1

Optimized enantiomeric separations of 100 chiral pesticides

Name1

# Isomers1

CSP2

Mobile phase3

F,T4

k 1 5a

α 5b

R s 5c

2-(2-Chlorophenoxy)propionic acid

2

QS10

PIM2

0.5, 25

2.1

1.15

2.4

TS10

PIM1b

1.0, 25

0.2

2.49

2.2

2-(3-Chlorophenoxy)propionic acid

2

QS10

PIM2

0.5, 25

1.5

1.15

2.1

TS10

PIM1b

1.0, 25

0.2

1.76

1.8

RSP10

RP2e

0.5, 25

2.2

1.09

1.5

2-(4-Chlorophenoxy)propionic acid

2

QS10

PIM2

0.5, 25

1.6

1.13

1.9

TS10

PIM1b

1.0, 25

0.2

3.13

2.5

RSP15

RP2f

0.8, 25

9.6

1.06

1.5

2-Phenoxypropionic acid

2

QS10

PIM2

0.5, 25

1.6

1.13

2.0

TS10

PIM1b

1.0, 25

0.2

2.39

2.1

Allethrin

8

RSP15

RP2a

0.8, 25

1.9

1.04

0.7

2.2

1.16

2.8

2.8

1.18

3.0

4.4

1.52

9.0

Anabasine

2

NS10

PIM1b

1.0, 45

3.0

1.16

2.8

VS10

POM1

1.0, 45

3.2

1.21

2.6

Anatabine

2

NS10

PIM1a

1.0, 45

1.1

1.45

5.2

VS10

RP1

1.0, 25

2.0

1.18

2.9

Ancymidol

2

RSP10

RP2b

0.5, 25

2.1

1.12

1.5

Benalaxyl

2

RSP10

RP2a

1.0, 25

0.6

1.36

2.4

Benoxacor

2

RSP15

RP2a

0.8, 25

0.6

1.31

2.8

Bifonazole

2

RSP15

RP1b

0.5, 25

7.1

1.08

1.4

Brodifacoum

4

QS10

PIM2

0.5, 25

2.2

1.17

2.5

4.2

1.78

9.0

Bromacil

2

VS10

RP1d

0.5, 25

2.8

1.15

1.9

TS10

RP1d

0.5, 25

2.0

1.20

2.5

Bromuconazole

4

RSP15

RP2a

0.8, 25

2.1

1.65

9.0

2.8

1.07

1.5

Butoconazole

2

RSP15

RP2c

0.5, 25

8.6

1.06

1.5

Carfentrazone

2

QS10

PIM2

0.5, 25

2.3

1.15

2.5

Carfentrazone-ethyl

2

QS10

RP1d

0.5, 25

1.4

1.33

2.0

RSP15

RP2a

0.8, 25

0.5

1.16

1.5

Chlorfenprop-methyl

2

RSP15

RP2a

0.8, 25

1.6

1.05

1.0

Chlorflurenol-methyl

2

RSP15

RP3c

0.5, 25

2.5

1.31

4.0

Cloquintocet-mexyl

2

RSP15

RP2

0.5, 25

0.9

1.21

2.0

Closantel

2

LSP10

NP3

0.8, 25

2.1

1.18

2.0

Coumachlor

2

NS10

NP4

1.0, 25

0.8

1.42

1.8

TS10

RP2a

1.0, 25

1.5

1.32

2.4

Coumatetralyl

2

QS10

PIM1c

1.0, 25

0.8

1.28

2.3

Cycloprothrin

4

RSP15

RP1b

0.8, 25

1.7

1.07

1.4

2.1

1.13

2.0

Cyfluthrin

8

RSP15

RP1b

0.5, 25

1.5

1.05

0.9

1.5

1.14

2.4

1.9

1.12

3.3

2.0

1.04

1.0

Cypermethrin

8

RSP15

RP1b

0.5, 25

3.2

1.02

0.3

3.2

1.29

1.9

3.6

1.07

0.7

4.2

1.07

0.9

Cyproconazole

4

RSP15

RP2c

0.5, 10

3.4

1.19

3.0

3.8

1.46

6.7

Dichlorprop

2

QS10

PIM1c

1.0, 25

0.3

1.35

2.3

TS10

PIM1b

1.0, 25

0.2

3.17

3.3

Diclofop

2

QS10

PIM2

0.5, 25

2.0

1.18

2.8

TS10

PIM1b

1.0, 25

0.2

2.93

4.0

Diclofop-methyl

2

RSP10

RP2a

0.7, 25

2.6

1.09

1.5

Diniconazole

2

QS10

RP1d

0.5, 25

2.5

1.14

1.5

RSP10

RP2b

0.5, 25

2.8

1.09

1.5

Dinoseb

2

RSP15

RP2d

0.8, 25

5.3

1.07

1.5

Dinotefuran

2

TS10

PIM1c

1.0, 25

0.6

1.13

1.0

Dyfonate

2

RSP15

RP2a

0.8, 25

2.7

1.04

0.9

Econazole

2

RSP10

RP2a

0.7, 25

0.9

1.17

1.6

Enilconazole

2

RSP10

RP2a

1.0, 25

0.5

1.26

1.9

EPN

2

RSP15

RP2a

1.0, 25

3.6

1.09

2.1

cis-Epoxiconazole

2

RSP10

RP2a

1.0, 25

0.9

1.19

1.8

Etaconazole

4

RSP15

RP1b

0.5, 25

1.6

1.21

4.7

2.0

1.15

2.8

Ethofumesate

2

RSP15

RP2e

1.0, 25

11.6

1.07

1.5

Etoxazole

2

RSP15

RP1b

0.5, 25

6.4

1.31

1.5

Famoxadone

2

LSP10

NP1

1.0, 25

7.0

1.39

4.5

RSP15

RP2a

1.0, 25

2.6

1.10

1.9

Fenamiphos

2

TS10

NP2

0.8, 25

1.7

1.14

2.0

Fenarimol

2

RSP15

RP2a

0.8, 25

2.5

1.04

0.9

Fenbuconazole

2

RSP15

RP4b

0.5, 25

6.9

1.04

0.8

Fenobucarb

2

RSP15

RP2d

0.8, 25

6.5

1.07

1.5

Fenoprop

2

TS10

RP1d

0.5, 45

1.0

1.29

1.9

Fenoxanil

4

RSP15

RP3c

0.8, 25

2.9

1.08

1.3

3.2

1.04

0.7

Fenoxaprop

2

QS10

PIM2

0.5, 25

2.0

1.12

2.0

TS10

PIM1b

1.0, 25

0.2

3.16

2.9

Fenoxaprop ethyl

2

RSP15

RP1b

0.5, 25

1.6

1.03

0.5

Fenpropathrin

2

RSP15

RP4a

0.5, 25

1.2

1.08

1.0

Fenvalerate

4

RSP15

RP1b

0.5, 25

2.0

1.04

0.9

2.2

1.12

2.5

Fipronil

2

LSP10

NP1

1.0, 25

5.3

1.24

3.7

Flucythrinate

4

RSP15

RP1b

0.5, 25

1.7

1.07

1.3

1.9

1.10

1.5

Fluroxypyr-1-methylheptyl ester

2

RSP10

RP2b

0.5, 25

1.6

1.15

1.5

Flurprimidol

2

RSP15

RP2a

1.0, 25

1.6

1.10

1.5

Flutriafol

2

RSP10

RP2a

0.5, 25

0.7

1.17

1.5

Furalaxyl

2

VS10

RP1d

0.5, 25

5.1

1.08

0.8

Haloxyfop

2

TS10

PIM1b

1.0, 25

0.1

4.03

2.4

Haloxyfop-methyl

2

RSP15

RP1b

0.5, 25

1.3

1.06

0.7

Hexaconazole

2

RSP10

RP2a

1.0, 25

1.1

1.34

2.3

Imazaquin

2

TS10

RP1e

0.5, 25

3.7

1.07

0.7

Isocarbophos

2

RSP15

RP3c

0.5, 25

1.9

1.11

1.6

Isofenphos-methyl

2

RSP10

RP2f

0.5, 25

3.9

1.06

0.9

cis-Ketoconazole

2

RSP15

RP2h

0.8, 25

2.5

1.11

1.5

Mandipropamid

2

TS10

RP1e

0.5, 45

2.8

1.10

1.7

Mecoprop

2

QS10

PIM2

0.5, 25

1.4

1.20

2.8

TS10

PIM1b

1.0, 25

0.1

2.88

2.4

Mecoprop methyl ester

2

NS10

RP1d

0.5, 25

6.8

1.17

2.2

VS15

RP1e

0.5, 45

3.2

1.07

1.5

Metconazole

4

RSP15

RP2h

0.8, 25

3.3

1.09

1.9

3.8

1.16

3.5

Miconazole

2

RSP10

RP2a

0.7, 25

0.9

1.17

1.7

Mitotane

2

RSP15

RP2a

0.8, 25

2.1

1.04

0.6

Myclobutanil

2

RSP15

RP1c

0.5, 25

2.2

1.06

1.2

Napropamide

2

VS10

RP1a

0.3, 25

0.1

1.86

1.5

TS10

RP1e

1.0, 45

2.4

1.14

1.8

Nicotine

2

NS10

PIM1b

1.5, 45

0.5

1.60

3.0

N-Methylanabasine

2

NS10

PIM2

1.0, 25

1.9

1.24

2.5

VS10

PIM1b

0.5, 25

1.3

1.38

3.2

Nornicotine

2

NS10

PIM1a

1.0, 45

2.7

1.14

2.3

VS10

PIM1d

0.7, 45

4.1

1.10

1.5

rac-(2R,3R)-Paclobutrazol

2

RSP15

RP2g

1.0, 25

3.2

1.09

1.8

Penconazole

2

RSP15

RP2b

0.8, 25

2.6

1.07

1.5

Pentanochlor

2

RSP15

RP3c

0.5, 25

8.1

1.04

0.6

Penthiopyrad

2

RSP15

RP3c

0.5, 25

4.4

1.13

2.1

Permethrin

4

RSP15

RP3a

0.4, 25

4.2

1.14

2.0

5.6

1.02

0.5

d-Phenothrin

2

RSP15

RP2a

0.8, 25

4.7

1.53

5.6

Phenthoate

2

RSP10

RP2b

0.5, 25

1.5

1.14

1.7

(1R;cis/trans;S)-Prallethrin

2

RSP15

RP2a

0.8, 25

2.2

1.57

9.0

Praziquantel

2

RSP15

RP2g

0.7, 45

2.3

1.11

1.5

TS10

RP5

0.8, 25

12.3

1.15

2.0

Propiconazole

4

RSP15

RP3b

0.5, 25

1.8

1.19

2.0

2.6

1.13

1.5

Prothioconazole

2

RSP10

RP2a

1.0, 25

1.5

1.18

2.2

Quizalofop ethyl

2

RSP15

RP2i

1.0, 25

7.4

1.05

0.9

Resmethrin

4

RSP15

RP4a

0.5, 25

2.4

1.04

0.7

3.1

1.31

4.5

Ruelene (Crufomate)

2

NS10

RP2b

1.0, 25

3.6

1.09

1.6

Spiroxamine

2

QS10

NP1

1.0, 25

0.3

1.54

1.5

Sulprofos

2

RSP15

RP1b

0.5, 25

2.3

1.08

1.8

Tebuconazole

2

RSP25

RP5

0.5, 25

4.4

1.07

1.5

Tetramethrin

4

RSP15

RP4a

0.5, 25

0.9

1.08

0.8

1.8

1.33

3.0

Tetramisole

2

NS10

PIM2

0.7, 25

3.5

1.13

2.1

Triadimefon

2

RSP15

RP2a

0.8, 25

1.4

1.09

1.5

Triadimenol

4

RSP15

RP2c

0.5, 25

1.5

1.44

5.7

2.3

1.06

1.6

Triticonazole

2

RSP15

RP2b

0.8, 25

4.3

1.09

1.5

(E)-Uniconazole

2

RSP15

RP2a

1.0, 25

2.2

1.08

1.5

Warfarin

2

NS10

RP2a

1.0, 25

4.4

1.11

1.7

TS10

RP2a

1.0, 25

1.1

1.29

2.3

VS10

RP2a

1.0, 25

4.3

1.19

2.0

1Pesticide name and number of isomers, see “Experimental” for more information

2Chiral stationary phase (CSP) with length in cm denoted in superscript [all 0.46 cm (i.d.)], 10: 10 cm × 0.46 cm (id), 15: 15 cm × 0.46 cm (id), 25: 10 × 0.46 cm (id) coupled to 15 × 0.46 cm (id). These include CDShell-RSP (RSP), NicoShell (NS), TeicoShell (TS), VancoShell (VS), LarihcShell-P (LSP) and Q-Shell (QS). See Experimental for more information

3See experimental for optimized mobile phases and acronyms

4Flow (F) in mL/min/temperature (T) in °C

5a,b,cCalculated chromatographic parameters of retention factor of the first peak (k1), selectivity (α), and resolution (Rs). See Experimental for more information

Data Processing

The dead time, \({t}_{0}\), was determined by the peak of the refractive index change due to the unretained sample solvent. Retention factors (k) were calculated using k = (tRt0)/(t0), where tR is the retention time of the first peak and t0, the dead time of the column. Selectivity (α) was calculated using α = k2/k1, where k1 and k2 are retention factors of the first and second peaks, respectively. Resolution (Rs) was calculated using the peak width at half peak height, Rs = 2(tR2tR1)/(w1 + w2). Iterative curve fitting was performed with PeakFit version 4.12 (SeaSolv Software Inc. 1999–2003). See Results and Discussion and Electronic Supplementary Material for the curve fitting protocol.

Results and Discussion

Core–shell CSPs were clearly applicable for the enantiomeric separation of chiral pesticides, as shown with all isomers of 74 pesticides baseline-separated (Rs > 1.5) with at least one CSP (Table 1). Further, 18 compounds were baseline-separated with two or more CSPs, and three compounds with 3 CSPs. Most pesticides required little optimization to provide Rs > 1.5, but different alcohols and salt concentrations were used to improve efficiency and/or selectivity. Most separations were performed in under ten minutes, and many compounds contained more than two stereoisomers. In Table 1, the selectivity and resolution are reported for each enantiomeric pair. The partial separations that could not be optimized to baseline without extensive optimization are not shown in Table 1. However, a partial separation with one chiral selector was often brought to baseline with one of the other related selectors, demonstrating the principle of complementary separations [31]. If there were one or more critical pairs that could not be baseline-separated without extensive optimization, the iterative curve fitting procedure was utilized. Overall, 69 pesticides were separated by CDShell-RSP, 19 by TeicoShell, 14 by Q-Shell, ten by NicoShell, nine by VancoShell, and three by LarihcShell-P. Representative chromatograms of each pesticide class separated by different CSPs are shown in Fig. 1, which will be discussed in subsequent sections.

Fig. 1

Optimized enantiomeric separations of pesticides from Table 1 indicative of their class. Enantiomeric separations of a cycloprothrin, b triadimenol, c EPN, d benalaxyl, e haloxyfop, f brodifacoum by ad CDShell-RSP, e TeicoShell, f Q-Shell. All conditions are listed in Table 1

Broad Spectrum of Enantioselectivity by CDShell-RSP

CDShell-RSP had the broadest selectivity, enough to achieve Rs > 1.5, for all pesticide classes and in the reversed phase mode. A representative chromatogram of the pyrethroid class of compounds is shown in Fig. 1a for cycloprothrin, which has four stereoisomers. Most pyrethroids, including permethrin, resmethrin, and tetramethrin produced similar results. Allethrin, which has eight stereoisomers, provided the best example of a critical overlapping pair. Two stereoisomers were overlapped, while the other six were baseline-separated. In the past, β-cyclodextrin has been used for the separation of allethrin but had much less resolution, which shows the power of the current core–shell CSPs [24]. Other pyrethroids with eight stereoisomers, like cyfluthrin and cypermethrin, were less resolved than allethrin, with more than one overlapping pair. The area extraction of these overlapped peaks will be discussed in a subsequent section and reported in Table 2.

Table 2

Area extraction of overlapped peaks using iterative curve fitting of 13 compounds

Name

# Stereo.

# Overlap1

Model2a

% Area2b

Std. error2c

R 2 2d

Allethrin

8

2

EMG

Peak1 = 5.8

0.774

0.9983

Peak2 = 5.6

Cyfluthrin

8

8

EMG

Peak1 = 11.4

1.913

0.9933

Peak2 = 18.2

Peak3 = 10.7

Peak4 = 17.9

Peak5 = 9.8

Peak6 = 10.9

Peak7 = 11.6

Peak8 = 9.5

Cypermethrin

8

8

EMG

Peak1 = 7.8

0.4748

0.9996

Peak2 = 18.3

Peak3 = 9.0

Peak4 = 12.1

Peak5 = 14.4

Peak6 = 18.0

Peak7 = 9.4

Peak8 = 10.9

Cycloprothrin

4

2

EMG

Peak1 = 25.8

0.122

0.9995

Peak2 = 26.0

Etaconazole

4

2

GMG

Peak2 = 21.6

0.276

0.9985

Peak3 = 28.5

Fenoxalil

4

4

EMG

Peak1 = 26.3

0.024

0.9993

Peak2 = 26.1

Peak3 = 25.0

Peak4 = 22.6

Fenpropathrin

2

2

EMG

Peak1 = 49.6

0.108

0.9994

Peak2 = 50.4

Fenvalerate

4

3

EMG

Peak1 = 26.1

0.338

0.9986

Peak2 = 23.2

Peak3 = 25.7

Flucythrinate

4

3

GMG

Peak1 = 26.1

1.514

0.9996

Peak2 = 24.8

Peak3 = 25.4

Metconazole

4

2

GMG

Peak2 = 41.9

0.069

0.9992

Peak3 = 9.1

Permethrin

4

2

GMG

Peak3 = 19.2

0.351

0.9980

Peak4 = 19.5

Resmethrin

4

2

GMG

Peak1 = 15.0

0.295

0.9967

Peak2 = 14.3

Tetramethrin

4

2

EMG

Peak1 = 11.1

0.250

0.9991

Peak2 = 11.3

1 Number of overlapped pairs from the chromatographic conditions in Table 1

2a,b,c,dSee “Results and discussion” and Supplemental Material for protocol using iterative curve fitting with the exponentially modified Gaussian model (EMG), Gaussian modified Gaussian (GMG) and a linear background using PeakFit software (2a). Extracted area percentages of overlapped peaks were determined from this protocol and numbered by their elution order in original chromatograms (see Table 1 for data) (2b). The standard error (std. error) and coefficient of determination (R2) of the fit were also obtained from PeakFit and reported (2c,2d)

The most powerful application of CDShell-RSP was for chiral fungicides (Fig. 1b). In Fig. 2, the universal nature of the CDShell-RSP for chiral fungicides was highlighted with the simultaneous LC–MS enantiomeric separation of 14 chiral fungicides in 20 min. This LC–MS approach would allow the simultaneous investigation of 32 fungicide enantiomers in foods, beverages, and any number of biological systems. These 32 enantiomers were from the four respective isomers of each bromuconazole and propiconazole, and 24 stereoisomers of the other fungicides (Fig. 2). Others with four stereoisomers which required different mobile phase compositions included cyproconazole and triadimenol (Table 1). Etaconazole and metconazole were baseline-separated, except for one overlapping pair (Table 2). CDShell-RSP also had high selectivity for the topical anti-fungal pesticides like miconazole and econazole (Fig. 2; Table 1). Other baseline-separated fungicides included the recently commercialized carboxamide, penthiopyrad, and the dicarboximide, famoxadone (Table 1).

Fig. 2

Simultaneous enantiomeric separation of 14 racemic fungicides using liquid chromatography electrospray ionization mass spectrometry (LC–ESI–MS). Total ion chromatogram (TIC) and extracted ion chromatograms (EICs) are shown. Conditions: CDShell-RSP, 150 × 4.6 mm (i.d.), ACN-NH4HCO2 (pH 3.6, 16 mM) (25:75), 0.8 mL/min, 25 °C. See “Experimental” for more information

CDShell-RSP also was selective for most organophosphates that had phosphorous stereogenic centers as shown in Fig. 1c. Some partially resolved organophosphates were ruelene (crufomate) and fenamiphos, which were baseline-separated by other core–shell CSPs (Table 1). Also, the acylanilide, benalaxyl was easily baseline-separated by CDShell-RSP (Fig. 1d), but others like mandipropamid and napropamide were best separated by other core–shell CSPs. An acaricide, etoxazole, and an imidazolidinone, bifonazole, were separated by CDShell-RSP (Table 1). Herbicides and their associated safeners [50], as well as other plant growth regulators without acidic functionalities, such as benoxacor, chlorfenprop-methyl, chlorflurenol-methyl, cloquintocet-mexyl, diclofop-methyl, and flupiridimol, were baseline-separated by CDShell-RSP. Herbicide metabolites, like the phenoxypropionic acid herbicides, were better separated by other core–shell CSPs, like TeicoShell and Q-Shell. Overall, these results indicate that CDShell-RSP had the broadest enantioselectivity for the tested pesticide classes. Complementary separations of a chiral herbicide and a rodenticide with two stereogenic centers by other core–shell CSPs are shown in Fig. 1e, f.

Specific Enantioselectivity by Other Core–Shell CSPs

Other SPP CSPs had more specific enantioselectivity for certain pesticide classes. TeicoShell and Q-Shell had the highest selectivity for the phenoxypropionic acid herbicides, like haloxyfop (Fig. 1e). This is not surprising as teicoplanin and quinine have amino groups, which can interact with acidic moieties of the analytes [32, 39]. Usually, these separations were performed in polar ionic mode, with methanol and a volatile salt. Phenoxypropionic acid herbicides were almost always separated with 1.5 < Rs < 4.0 within 1–2 min using TeicoShell. Under environmental conditions, these herbicides are formed as metabolites via hydrolysis from phenoxypropionate herbicide derivatives [10]. The phenoxypropionate herbicides, like mecoprop methyl ester and carfentrazone-ethyl, were found to separate better by other CSPs, including NicoShell and CDShell-RSP (Fig. 3).

Fig. 3

Enantiomeric separation trends of phenoxypropionate herbicides and their phenoxypropionic acid metabolites. a Enantiomeric separations of mecoprop, a phenoxypropionic metabolite of (b) mecoprop methyl ester using TeicoShell and NicoShell. c Enantiomeric separations of carfentrazone, a triazolone herbicide metabolite of (d) carfentrazone-ethyl, using Q-Shell and CDShell-RSP. See Table 1 for conditions

Another class of pesticides with high selectivity using TeicoShell or Q-Shell were the rodenticides (Fig. 1f). First-generation anticoagulants, coumachlor and warfarin, were best separated using macrocyclic glycopeptide-based CSPs like TeicoShell, as expected (Table 1) [23, 39, 40]. Brodifacoum, a newer anticoagulant meant for larger rodents, is a much larger molecule than older generation rodenticides and has 4 stereoisomers. Brodifacoum had no selectivity with TeicoShell and was partially separated by CDShell-RSP. It was best separated by Q-Shell as shown in Fig. 1f. Q-Shell was also the only CSP to baseline-separate coumatetralyl (Table 1). However, Q-Shell had no selectivity for coumachlor and warfarin. Extensive enantiomeric methods have not been published with Q-Shell, but it seems that it has high selectivity for phenoxypropionic acids, and complementary behavior with TeicoShell for larger rodenticides.

The other macrocyclic glycopeptide-based CSPs, VancoShell and NicoShell, were most useful to separate chiral pesticides with amine functionalities, like the tobacco-based insecticides (Table 1). NicoShell was the only CSP that baseline-separated the anti-fungal tetramisole and the organophosphate ruelene (crufomate) (Table 1). VancoShell was the only CSP that had selectivity for the acylanilide furalaxyl. A partial separation, which was only achieved by TeicoShell, was the neonicotinoid dinotefuran (Table 1). Since neonicotinoids have extreme toxicity towards honey bees, perhaps TeicoShell should be further evaluated for other chiral neonicotinoid separations, especially those with undetermined enantiomeric properties [14].

Perhaps, the most interesting pesticide separations were those by the isopropyl-cyclofructan 6, LarihcShell-P. This derivatized cyclofructan is best known for separating primary amines, which explains the separation of fipronil [31, 41]. However, it has also been reported to separate other non-primary amines, like Tröger’s base [41]. Further evidence of non-primary amine enantiomeric selectivity was observed with the separations of closantel and famoxadone (Table 1). These separations were only achieved in normal phase solvents, but no other CSP could baseline-separate closantel (Table 1). This shows that this chiral selector has broader enantioselectivity, not just for primary amines but also for neutral compounds.

Area Extraction of Overlapped Peaks Using Iterative Curve Fitting

Out of the 100 pesticides, 26 were partially resolved, but many like cycloprothrin (Fig. 1a) had multiple enantiomers that were all baseline-separated except for one pair of overlapping peaks. Instead of spending extensive time working on method development, iterative curve fitting was used to determine their overlapping peak areas (see Table 2). The subsequent protocol was followed based on previous work and is illustrated with two examples, then used for 11 other pesticides to evaluate its applicability (see Fig. 4; Table 2) [43]. First, the raw data of the best separation possible was obtained. The data was then imported to PeakFit software and smoothed with a Savitsky-Golay filter. Next, the model was chosen to fit the number of peaks expected with the chromatogram. Last, the curve fitting program simulated the fit and provided the areas of each peak.

Fig. 4

Iterative curve fitting of two pyrethroids. a Enantiomeric separation of allethrin with CDShell-RSP resulting in one overlapped enantiomeric pair. b Enantiomeric separation of cyfluthrin with CDShell-RSP resulting in several overlapping peaks. Areas of the overlapped peaks were extracted according to the protocol discussed in the “Results and Discussion” and Supplemental material. See Table 1 for chromatographic information, and Table 2 for extracted areas and other results

Using this simple four-step protocol, worked examples are shown for a simple case of one overlapping pair, and a more complex case with several overlapping peaks (Fig. 4, Fig. S1 in Electronic Supplementary Material). The eight isomers of the pyrethroid, allethrin, were all baseline-separated by CDShell-RSP except the first two peaks, which was representative of most pyrethroid enantiomeric separations (Fig. 4a). After using the protocol above, the extracted area percentages of peak 1 and peak 2 were determined after 68 iterations as 5.77 and 5.61%, respectively (Fig. 4a; Table 2). The number of peaks is chosen by the user, which might be erroneous with complex samples and complete overlap (Rs = 0.0), but with the MS-compatible methods reported in this study, quick MS peak purity checks can be performed. Hidden peaks with the same m/z values, like enantiomers, must be separated or their retention times determined separately as they cannot be differentiated by MS [51]. However, all enantiomers were visible for the standard compounds used in this study. Overall, this approach provided a better estimation of the two allethrin peak areas compared to simple integration methods (perpendicular drop), which reported the incorrect area percentages as 4.87 and 6.43%.

More difficult situations, as in the case of cyfluthrin, which had seven out of eight overlapping isomers on the CDShell-RSP, were also assessed by the iterative curve fitting protocol (Fig. 4b; Tables 1, 2). Since all the peaks were less resolved in comparison to allethrin, the curve fitting was more challenging, but with the exponentially modified Gaussian (EMG) model and seven iterations, the recovered areas were estimated (following elution order of peak 1 to peak 8) as 11.4%, 18.2%, 10.7%, 17.9%, 9.8%, 10.9%, 11.6%, and 9.5%. With ideal peak shapes and in the absence of noise, iterative curve fitting error has been reported as < 1% error [47]. The extensive overlap between these peaks makes simple integration very difficult, especially to determine which peaks are enantiomeric pairs. However, from this approach, the enantiomeric pairs can be better estimated as peaks 1 and 3, 2 and 4, 5 and 8, and 6 and 7. Further confirmation was provided with the separation of β-cyfluthrin, which only contains 4 of the 8 stereoisomers of cyfluthrin. These 2 enantiomeric pairs were aligned with peaks 1 and 3, and 5 and 8.

As seen in Fig. 4a, b, the area ratios between all the peaks were not equal, which might lead to errors if the peaks overlap when using conventional integration methods [42]. Iterative curve fitting is very flexible as even asymmetric peak shapes can also be modeled with high accuracy [45]. However, it is subjective to the user and which model is chosen. In the 12 separations assessed, the best models with least error seemed to be the exponentially modified Gaussian (EMG) and the Gaussian modified Gaussian (GMG) (Table 2). Overall, the peak areas were conveniently extracted from partially resolved chromatograms, even the pyrethroids with up to 8 stereoisomers. Undoubtedly, this procedure applies to all complex enantiomeric separations that require extensive chromatographic method development. It could provide quantitative information from overlapping signals, which might be useful for complex biological enantiomeric studies, especially those of pesticides. Further studies will determine if this approach is suitable for quantitative determination of trace components (< 1%).

Concluding Remarks

Effective methods with core–shell (superficially porous particles) CSPs and mass spectrometry-compatible mobile phases were established for 100 chiral pesticides, many with more than one chiral center. In past studies, cyclodextrins have had success in separating pesticides, but never has such a comprehensive study, with the evaluation of several chiral selectors towards chiral pesticides been reported. The hydroxypropyl-β-cyclodextrin CSP (CDShell-RSP) provided the broadest enantiomeric selectivity for chiral pesticide separations. High enantiomeric selectivity for pesticides with acidic functionalities were dominantly provided by the macrocyclic glycopeptide and quinine-based CSPs (TeicoShell and Q-Shell). Pesticides with amine functionalities and a few unique cases were better suited towards the other macrocyclic glycopeptides and derivatized cyclofructan (NicoShell, VancoShell, and LarihcShell-P). A convenient protocol using an iterative curve fitting method was developed and can be applied to any partially separated pesticides, which would provide the necessary information needed for biological and/or environmental enantiomeric studies.

Notes

Acknowledgements

We thank AZYP, LLC, for their technical support for HPLC chiral column technology. We also thank Siqi Du for her MS expertise. This work was supported by the Robert A. Welch Foundation (Y0026).

Compliance with Ethical Standards

Conflict of Interest

The authors J.T. Lee, D.A. Lopez, and D.W. Armstrong declare the following competing financial interest(s): CDShell-RSP, NicoShell, LarihcShell-P, Q-Shell, TeicoShell, and VancoShell are trademarked products of AZYP, LLC.

Ethical Approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Supplementary material

10337_2018_3604_MOESM1_ESM.pdf (387 kb)
Supplementary material 1 (PDF 386 KB)

References

  1. 1.
    IUPAC (1997) Compendium of Chemical Terminology, 2nd edn. (the “Gold Book”). Blackwell Scientific Publications, OxfordGoogle Scholar
  2. 2.
    Matell M (1953) Ark Kemi 6:365–373Google Scholar
  3. 3.
    Schneiderheinze JM, Armstrong DW, Berthod A (1999) Chirality 11:330–337CrossRefGoogle Scholar
  4. 4.
    Burden RS, Carter GA, Clark T, Cooke DT, Croker SJ, Deas AHB, Hedden P, James CS, Lenton JR (1987) Pestic Sci 21:253–267CrossRefGoogle Scholar
  5. 5.
    Calcaterra A, D’Acquarica I (2018) J Pharm Biomed Anal 147:323–340CrossRefGoogle Scholar
  6. 6.
    Armstrong DW, Chang CD, Li WY (1990) J Agric Food Chem 38:1674–1677CrossRefGoogle Scholar
  7. 7.
    Armstrong DW, Reid GL III, Hilton ML, Chang CD (1993) Environ Pollut 79:51–58CrossRefGoogle Scholar
  8. 8.
    Liu W, Gan J, Schlenk D, Jury WA (2005) Proc Natl Acad Sci USA 102:701–706Google Scholar
  9. 9.
    Ye J, Zhao M, Liu J, Liu W (2010) Environ Pollut 158:2371–2383CrossRefGoogle Scholar
  10. 10.
    Perez de Albuquerque NC, Carrão DB, Habenschus MD, Moraes de Oliveira AR (2018) J Pharm Biomed Anal 147:89–109CrossRefGoogle Scholar
  11. 11.
    Schleier JJ III, Peterson RKD (2011) Pyrethrins and pyrethroid insecticides. In: Lopez Ó, Fernández-Bolaños JG (eds) Green Trends in Insect Control, RSC Green Chemistry, No. 11. RSC Publishing, Cambridge, pp 94–131CrossRefGoogle Scholar
  12. 12.
    Miyamoto J (1976) Environ Health Perspect 14:15–28CrossRefGoogle Scholar
  13. 13.
    Ma Y, Chen L, Lu X, Chu H, Xu C, Liu W (2009) Ecotoxicol Environ Saf 72:1913–1918CrossRefGoogle Scholar
  14. 14.
    Oberhauser KS, Brinda SJ, Weaver S, Moon RD, Manweiler SA, Read N (2006) Environ Entomol 35:1626–1634CrossRefGoogle Scholar
  15. 15.
    Johnson RM, Ellis MD, Mullin CA, Frazier M (2010) Apidologie 441:312–331CrossRefGoogle Scholar
  16. 16.
    Vontas J, Grigoraki L, Morgan J, Tsakireli D, Fuseini G, Segura L, Niemczura de Carvalho J, Nguema R, Weetman D, Slotman MA, Hemingway J (2018) Proc Natl Acad Sci USA 115(18):4619–4624Google Scholar
  17. 17.
    Chen ZM, Wang YH (1996) J Chromatogr A 754:367–395CrossRefGoogle Scholar
  18. 18.
    Alder L, Greulich K, Kempe G, Vieth B (2006) Mass Spectrom Rev 25:838–865CrossRefGoogle Scholar
  19. 19.
    Ye J, Wu J, Liu W (2009) Trends Anal Chem 28:1148–1163CrossRefGoogle Scholar
  20. 20.
    Okamoto M (2012) Direct chiral separation of pyrethroid isomers by HPLC with chiral stationary phases In: Knaak JB, Timchalk C, Tornero-Velez R (eds) Parameters for Pesticide QSAR and PBPK/PD Models for Human Risk Assessment, ACS Symposium Series, vol 1099, pp 31–40Google Scholar
  21. 21.
    Ȏi N, Kitahara H, Kira R (1990) J Chromatogr A 515:441–450CrossRefGoogle Scholar
  22. 22.
    Lisseter SG, Hambling SG (1991) J Chromatogr A 539:207–210CrossRefGoogle Scholar
  23. 23.
    Armstrong DW, Chang CD, Lee SH (1991) J Chromatogr A 539:83–90CrossRefGoogle Scholar
  24. 24.
    Kutter JP, Class TJ (1992) Chromatographia 33:103–112CrossRefGoogle Scholar
  25. 25.
    Zhang H, Qian M, Wang X, Wang X, Xu H, Wang Q, Wang M (2012) J Sep Sci 35:773–781CrossRefGoogle Scholar
  26. 26.
    Li Y, Dong F, Liu X, Xu J, Li J, Kong Z, Chen X, Liang X, Zheng Y (2012) J Chrom A 1224:51–60CrossRefGoogle Scholar
  27. 27.
    Patel DC, Wahab MF, Armstrong DW, Breitbach ZS (2016) J Chromatogr A 1467:2–18CrossRefGoogle Scholar
  28. 28.
    Hellinghausen G, Lee JT, Weatherly CA, Lopez DA, Armstrong DW (2017) Drug Test Anal 9:944–948CrossRefGoogle Scholar
  29. 29.
    Hellinghausen G, Roy D, Wang Y, Lee JT, Lopez DA, Weatherly CA, Armstrong DW (2017) Talanta 181:132–141CrossRefGoogle Scholar
  30. 30.
    Barhate CL, Lopez DA, Makarov AA, Bu X, Morris WJ, Lekhal A, Hartman R, Armstrong DW, Regalado EL (2018) J Chrom A 1539:87–92CrossRefGoogle Scholar
  31. 31.
    Hellinghausen G, Roy D, Lee JT, Wang Y, Weatherly CA, Lopez DA, Nguyen KA, Armstrong JD, Armstrong DW (2018) J Pharm Biomed Anal 155:70–81CrossRefGoogle Scholar
  32. 32.
    Patel DC, Breitbach ZS, Yu J, Nguyen KA, Armstrong DW (2017) Anal Chim Acta 963:164–174CrossRefGoogle Scholar
  33. 33.
    Gritti F, Guiochon G (2012) LC GC N Am 30:586–595Google Scholar
  34. 34.
    Broeckhoven K, Cabooter D, Desmet G (2013) J Pharm Anal 3:313–323CrossRefGoogle Scholar
  35. 35.
    DeStefano JJ, Schuster SA, Lawhorn JM, Kirkland JJ (2012) J Chromatogr A 1258:76–83CrossRefGoogle Scholar
  36. 36.
    Kotoni D, Ciogli A, Molinaro C, D’Acquarica I, Kocergin J, Szczerba T, Ritchie H, Villani C, Gasparrini F (2012) Anal Chem 84:6805–6813CrossRefGoogle Scholar
  37. 37.
    Ciogli A, Ismail OH, Mazzoccanti G, Villani C, Gasparrini F (2018) J Sep Sci 41:1307–1318CrossRefGoogle Scholar
  38. 38.
    Xu C, Armstrong DW (2013) Anal Chim Acta 792:1–9CrossRefGoogle Scholar
  39. 39.
    Armstrong DW, Liu Y, Ekborg-Ott KH (1995) Chirality 7:474–497CrossRefGoogle Scholar
  40. 40.
    Armstrong DW, Tang Y, Chen S, Zhou Y, Bagwill C, Chen JR (1994) Anal Chem 66:1473–1484CrossRefGoogle Scholar
  41. 41.
    Sun P, Wang C, Breitbach ZS, Zhang Y, Armstrong DW (2009) Anal Chem 81:10215–10226CrossRefGoogle Scholar
  42. 42.
    Bicking MKL (2006) LC GC N Am 24:605–616Google Scholar
  43. 43.
    Chesler SN, Cram SP (1973) Anal Chem 45:1354–1359CrossRefGoogle Scholar
  44. 44.
    Anderson AH, Gibb TC, Littlewood AB (1970) Anal Chem 42:434–440CrossRefGoogle Scholar
  45. 45.
    Amigo JM, Skov T, Bro R (2010) Chem Rev 110:4582–4605CrossRefGoogle Scholar
  46. 46.
    Kemmer G, Keller S (2010) Nat Protoc 5:267–281CrossRefGoogle Scholar
  47. 47.
    Asher BJ, D’Agostino LA, Way JD, Wong CS, Harynuk JJ (2009) Chemosphere 75:1042–1048CrossRefGoogle Scholar
  48. 48.
    Wahab MF, Wimalasinghe RM, Wang Y, Barhate CL, Patel DC, Armstrong DW (2016) Anal Chem 88:8821–8826CrossRefGoogle Scholar
  49. 49.
    Patel DC, Wahab MF, O’Haver TC, Armstrong DW (2018) Anal Chem 90:3349–3356CrossRefGoogle Scholar
  50. 50.
    Kraehmer H, Laber B, Rosinger C, Schulz A (2014) Plant Physiol 166:1119–1131CrossRefGoogle Scholar
  51. 51.
    Boehm RE, Martire DE, Armstrong DW (1988) Anal Chem 60:522–528CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of Chemistry and BiochemistryThe University of Texas at ArlingtonArlingtonUSA
  2. 2.AZYP LLCArlingtonUSA

Personalised recommendations