gas-phase fragmentation of protonated benzodiazepines

RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2007; 21: 2273–2281

) DOI: 10.1002/rcm.3084
Published online in Wiley InterScience (www.interscience.wiley.com
Gas-phase fragmentation of protonated benzodiazepines

Antonella Risoli1,2, Joey B. Y. Cheng1, Udo H. Verkerk1, Junfang Zhao1, Gaetano Ragno2,

Alan C. Hopkinson1 and K. W. Michael Siu1*1Department of Chemistry and Centre for Research in Mass Spectrometry, York University, 4700 Keele Street, Toronto, Ontario,

Canada M3J 1P32Department of Pharmaceutical Sciences, University of Calabria, 87036 Arcavacata di Rende, Cosenza, Italy

Received 5 April 2007; Revised 7 May 2007; Accepted 8 May 2007

*CorrespoCentre foKeele StrE-mail: kContract/Research

Protonated 1,4-benzodiazepines dissociate in the gas phase by the common pathway of CO eli-

mination and by unique pathways dictated by the substituents; the latter typically differentiate

one benzodiazepine from another. Protonated 3-dihydro-5-phenyl-1,4-benzodiazepin-2-one, the

base diazepam devoid of substituents, dissociates by eliminating CO, HNCO, benzene, and ben-

zonitrile. Mechanisms of these reactions are proposed with ionic products being resonance stabil-

ized. The abundant [MHSCO]R ion dissociates to secondary products via elimination of benzene,

benzonitrile, the NH2 radical, and ammonia, yielding again ionic products that are stabilized by

resonance. Copyright # 2007 John Wiley & Sons, Ltd.

Benzodiazepines were discovered serendipitously in 1957

in the Roche laboratories.1 The benzodiazepine class of

drugs consists of approximately 50 commercially available

compounds with anxiolytic, muscle-relaxant and sedative

effects.2,3 In clinical practice these drugs are widely pres-

cribed for the treatment of insomnia and anxiety disorders.

Benzodiazepines are central nervous system (CNS) depress-

ants4,5 and are characterized by a strong and specific inter-

action with local g-aminobutyric acid chloride channel

receptor (GABAA) sites.6–10 Various benzodiazepine deriva-

tives were originally marketed as an improvement over the

then available barbiturates with regard to overdose effects,

drug interactions, dependence development, and illicit

abuse. Over-prescription in the 1970s and 1980s revealed

side effects and led to illicit use; this eventually resulted in a

decline in benzodiazepine prescriptions.3 Use of benzo-

diazepines can lead to physiological and psychological

withdrawal symptoms upon discontinuation;11,12 when

taken in combination with other CNS depressants, includ-

ing alcohol, benzodiazepines can induce severe, even fatal,

respiratory failure.13–15

The determination of benzodiazepines is often required

in forensic and clinical analyses. The large number of

benzodiazepines, together with the multiple metabolites of

varying biological activity that each of these will form,2,3,12,16

poses a significant analytical challenge.12,13,17 Currently,

gas chromatography (GC)18 and liquid chromatography

(LC)17,19–22 coupled with mass spectrometry (MS) are the

most commonly employed methods. MS has superior selec-

tivity and sensitivity than other detection methods; MS

evidence is considered to be the most reliable and less likely

ndence to: K. W. M. Siu, Department of Chemistry andr Research inMass Spectrometry, York University, 4700eet, Toronto, Ontario, Canada M3J [email protected] sponsor: Natural Sciences and EngineeringCouncil (NSERC) of Canada and MDS SCIEX.

to be challenged in court.12,23 While electron ionization is the

major ionization technique in GC/MS,24–27 softer ionization

techniques, including electrospray ionization (ESI), have

been widely used in LC/MS in recent years. For these softer

ionization techniques, structural information on the benzo-

diazepines can be derived from tandem mass spectrometry

(MS/MS). Qualitative analysis has been demonstrated using

an MS/MS library acquired under specific experimental

conditions.28,29

In general, the collision-induced dissociation (CID) of

benzodiazepines results in rich MS/MS spectra.11,17,28–34

Typically, the cleavage of characteristic substituents, includ-

ing Cl, NO2 and OH, coupled with GC or LC retention times

and molecular weights discernible from ESI, can result in the

(near) unambiguous identification of existing benzo-

diazepines. The present study was driven by a desire to

better understand the CID fragmentation chemistries in

order to enhance the ability to assign structural components,

when a new benzodiazepine is encountered. Towards this

end, we first examine the intrinsic dissociation reactions

of benzodiazepines by analyzing the CID of a skeletal

benzodiazepine, 3-dihydro-5-phenyl-1,4-benzodiazepin-2-

one (see Scheme 1 for the structure), that is devoid of any

functional group which typically differentiates the various

benzodiazepine products. For simplicity, this benzo-

diazepine will be referred hereafter as the base diazepam.

Characteristic fragmentations of benzodiazepine products

will then be introduced and discussed with respect to

features that will aid identification of these products by

recognizing the functional groups or substituents that they

carry.

Copyright # 2007 John Wiley & Sons, Ltd.

Scheme 1.

2274 A. Risoli et al.

EXPERIMENTAL

MaterialsNitrazepam, norfludiazepam, lorazepam (see Scheme 1 for

structures and molecular weights) were obtained from

Cambridge Isotope Laboratories (Quebec, Canada) as

solutions at a certified concentration of 1mg/mL in

methanol. All other benzodiazepines, including base diaze-

pam, nordiazepam, clonazepam, diazepam, oxazepam,

temazepam, and flunitrazepam (Scheme 1), at 10 ng/mL in

water/methanol, were a gift from MDS SCIEX (Concord,

ON, Canada). Sample solutions for MS analysis were at a

concentration of 4mM in 50:50 water/methanol containing

1% acetic acid. Methanol and acetic acid were from

Sigma (St. Louis, MO, USA); water was purified using a

Milli-Q system (Millipore, Billerica, MA, USA). All diazepam

materials were used as received and stored at 48C except for

analysis.

Mass spectrometryExperiments were conducted on a pre-commercial version of

the QTRAP linear ion-trap mass spectrometer (Applied

Biosystem/MDS SCIEX). Sample solutions were infused

into the pneumatically assisted ESI source at a flow rate


of 2–3mL/min using a syringe pump. The typical ESI voltage

was between 5 and 5.5 kV. Dry air was used as the nebuli-

zer gas. The orifice and lens voltages were optimized

for maximum precursor ion signal intensity and between

100 and 200 scans were summed to produce a mass

spectrum. In a typical MS/MS experiment, the precursor

ion was mass-selected using the first quadrupole (Q1),

allowed to collide with nitrogen at a pressure of approxi-

mately 7.5mTorr (CAD gas setting of 4) in the second

quadrupole (q2), and mass-analyzed using the third

quadrupole (Q3). For chlorine-containing benzodiazepines,

unless otherwise stated, the reported results are based on

analyses of the 35Cl-containing ions, although the 35Cl- and

the 37Cl- containing ions were both examined to verify ion

assignments. For CID, the laboratory collision energy (Elab)

typically varied between 5 and 50 eV (100 eV for the base

diazepam). The ion lineage of specific product ions was

further investigated using the enhanced product ion (EPI)

scan mode, which traps and accumulates product ions in Q3

for better sensitivity.

Computational methodsGeometry optimizations and energy calculations were

performed with Gaussian 0335 using the non-local hybrid


DOI: 10.1002/rcm

Figure 1. Product ion spectrum of protonated base diaze-

pam, m/z 237, at Elab¼ 30 eV.

Gas-phase fragmentation of protonated benzodiazepines 2275

three parameter B3LYP density functional theory (DFT)

approach36,37 with the 6-31þG(d) split-valence basis sets. All

stationary points were characterized by harmonic frequen-

cies as local minima (all real frequencies) and saddle points

(one imaginary frequency). Connections between transition

states and corresponding minima were verified using the

intrinsic reaction coordinate method.38,39

RESULTS AND DISCUSSION

A benzodiazepine has three heteroatoms on the seven-

membered ring and each potentially can accommodate

the ‘ionizing’ proton to give a protonated benzodiazepine

(MHþ). These are the imine nitrogen at N4, the amide oxy-

gen attached to C2, and the amide nitrogen at N1 (Scheme 1),

Figure 2. (a) Energy-resolved CID results of MHþ, m/z 237 and (b) details of secondary

products of the [MH�CO]þ ion, m/z 209.

Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 2273–2281

DOI: 10.1002/rcm

Figure 3. Energy profile for the elimination of CO from protonated base diazepam at B3LYP/6-31þG(d).


in decreasing order of basicity. DFT calculations show that

protonation on the imine nitrogen produces ion I that is 23.1

and 36.9 kcal/mol lower in enthalpy (DH80) than the ions

produced by protonation on the amide oxygen, II, and on the

amide nitrogen, III, respectively (Scheme 2).

Figure 1 shows the CID spectrum (product ion scan) of

protonated base diazepam at a laboratory collision energy

(Elab) of 30 eV. Energy-resolved CID results are shown in

Fig. 2 for the MHþ (m/z 237) as well as details of secondary

products from a prominent product ion at m/z 209, formed

by elimination of CO. The loss of CO is a feature shared

by all benzodiazepines.11,17,28–34 Figure 3 gives the energy

profiles for the loss of CO from ion I directly and indi-

rectly via ion III. The former has a higher enthalpy barrier

(DH80¼ 73.3 kcal/mol) than the latter (56.3 kcal/mol). Pro-

tonation on the amide nitrogen weakens the amide bond

Scheme


(a fact well known in the CID of protonated peptides).40 It is,

therefore, not surprising that the reaction pathway involving

ion III has a lower energy barrier. The barrier against

conversion of ion I into ion III, involving a 1,4-proton shift, is

54.2 kcal/mol – slightly lower than the barrier against the loss

of CO from III.

In addition to the loss of CO, protonated base diazepam

can also lose neutrals of 43, 78, and 103Da to give relatively

minor product ions at m/z 194, 159, and 134, respectively

(Figs. 1 and 2(a)). Plausible mechanisms are shown in

Scheme 3. In pathway B, ring contraction leads to the loss of

HNCO, 43Da. The resulting ion at m/z 194 delocalizes its

positive charge onto the original N4 (IV), the benzylic C5,

and both benzene rings. In pathwayC, a 1,3-proton shift from

N4 to C10 on the phenyl ring induces elimination of the

phenyl ring as benzene, 78Da, giving a benzodiazepine ion at

2.


DOI: 10.1002/rcm

Scheme 3.


m/z 159 (V). By contrast, pathway D involves a 1,4-proton

shift that results in protonation of the amide oxygen to give

ion II. Contraction of the diazepine ring leads to elimination

of benzonitrile, 103Da, and formation of protonated oxindole

(m/z 134), ion VI. Alternatively, elimination of oxindole from

I by ring contraction would produce protonated benzonitrile

and a subsequent proton transfer would yield VI.

The majority of product ions seen in Fig. 1 are second-

ary product ions (Fig. 2(b)) as a result of dissociation of

the primary product ion, amino-protonated 1,2-dihydro-4-

phenylquinazoline, ion P atm/z 209. Ion P(, imino-protonated

1,2-dihydro-4-phenylquinazoline, is at the global minimum,

23.7 kcal/mol below P, but the barrier against interconver-

sion between these tautomers is 33.8 kcal/mol above P. From


the energy-resolved results shown in Fig. 2(b), and ignoring

product ions with abundances <5%, it is apparent that the

following secondary product ions,m/z 131, 106, 193, and 192,

are formed directly from the [MH�CO]þ ion at m/z 209, and

with relatively low barriers. Mechanisms that can lead to

these ions are shown in Scheme 4. In pathway E, a 1,5-proton

shift in P fromN1 to C10 or a 1,3-proton shift in P( fromN4 to

C10 induces elimination of the phenyl ring as benzene and

gives the protonated 2-methaniminobenzonitrile ion (VII) at

m/z 131. Ring contraction in pathway F can occur from either

P or P( forming either protonated benzoazetine (ion VIII) at

m/z 106, or protonated benzonitrile at m/z 104. Ion VIII is

16.9 kcal/mol higher in enthalpy (DH80) than protonated

N-methylenebenzenamine, Ph�NHþ¼CH2; however, con-

version from the former into the latter is a high-energy

process, requiring a 1,3-proton shift and cleavage of a C–C

bond. The calculated proton affinities (PAs) of 1,2-dihydro-

4-phenylquinazoline (that protonates to give VIII) and

benzonitrile are 218.2 and 194.4 kcal/mol at B3LYP/6-31þG(d); the latter is in good agreement with the reference PA of

194 kcal/mol for benzonitrile.41 The observation of proto-

nated benzonitrile being formed at higher threshold energies

suggests that the dominant pathway is via ion P(. At low

collision energies, the lifetime of the initially formed

PhCNHþ/benzoazetine complex will be sufficiently long

to permit efficient proton transfer and protonated benzoa-

zetine will be the major product; at higher collision energies,

the lifetime of the complex will be shorter and direct

dissociation yields protonated benzonitrile. Pathways G and

H result in losses of the NH2 radical and NH3 leading to the

product ions atm/z 193 and 192, respectively. The initial step

is identical and involves cleavage of the N1–C2 bond in P,

giving IX, an ion in which there is extensive charge

delocalization. Elimination of NH3 (pathway G) yields a

benzyne ion X at m/z 192 that is similarly resonance

stabilized. Alternatively, elimination of �NH2 (pathway H)

results in a distonic phenyl radical XI at m/z 193 that is also

resonance stabilized.

Figure 4 shows the CID results of (a) protonated

nordiazepam and (b) protonated norfludiazepam, both

at Elab of 30 eV. Both benzodiazepines are simple derivatives

of the base diazepam: nordiazepam has a Cl substituent at

the C7 position, a rather common substitution pattern among

commercially available diazepams, while norfludiazepam

carries, in addition to the Cl substituent, a F substituent at the

C20 position, which is less common than the former

(Scheme 1). As observed for protonated base diazepam,

protonated nordiazepam (m/z 271) dissociates under CID

conditions by losing the following neutrals: CO, HNCO, and

benzonitrile. The product ion at m/z 193 is interpreted as

being formed via concomitant elimination of HNCO and Cl,

not benzene. This interpretation is a consequence of the

following observations: CID of the 37Cl-containing proto-

nated nordiazepam atm/z 273 yields a product ion atm/z 193;

and protonated norfludiazepam shows the same loss of 78Da

(HNCOþCl, Fig. 3(b)) and none at 96Da (the mass of

fluorobenzene). In addition, this interpretation is in agree-

ment with a previous one based on H/D-exchange.31

Protonated norfludiazepam (m/z 289) dissociates analo-

gously, via losses of CO (28Da), ClþHNCO (78Da), and


DOI: 10.1002/rcm

Scheme 4.


fluorobenzonitrile (121Da). The [MH�CO]þ ions of both

diazepam derivatives dissociate facilely to give secondary

product ions, as was observed for the [MH�CO]þ ion of the

base diazepam. For nordiazepam, the losses are Cl, benzene,

and benzonitrile; for norfludiazepam, they are Cl, fluoro-


benzene, and fluorobenzonitrile. The resulting ions should

have structures analogous to those postulated in Scheme 4.

Substituting the N1 hydrogen of nordiazepam with a

methyl group results in diazepam; the CID spectrum of

protonated diazepam at Elab of 30 eV is shown in Fig. 5. It


DOI: 10.1002/rcm

Figure 4. Product ion spectra of (a) protonated nordiaze-

pam, m/z 271 and (b) protonated norfludiazepam, m/z 289,

both at Elab¼ 30 eV.

Figure 6. Product ion spectra of (a) protonated nitrazepam,

m/z 282, at Elab¼ 30 eVand (b) protonated flunitrazepam,m/z

314, at Elab¼ 35 eV.


shows many similarities to the CID spectrum of protonated

nordiazepam in Fig. 2(a). The two new features are the

elimination of neutrals of 57Da from protonated diazepam

and of 29Da from the [MH�CO]þ ion. The 57Da neutral that

is lost is interpreted as CH3NCO; the mechanism for its

elimination is postulated to be analogous to that of HNCO

shown in Scheme 3. Protonated nordiazepam does not

appear to lose HNCO alone, but concomitantly with Cl. The

29Da neutral lost from the [MH�CO]þ ion corresponds to

methyl nitrene, CH3N (or the lower energy H2C¼NH),

forming protonated isoindole.

Substituting the C7 hydrogen on nordiazepamwith a nitro

group results in nitrazepam; the CID spectrum of protonated

nitrazepam at Elab of 30 eV is shown in Fig. 6(a). The most

prominent feature is the loss of NO2, 46Da. This same feature

Figure 5. Product ion spectrum of protonated diazepam,m/z

286, at Elab¼ 30 eV.


is evident in the CID of protonated flunitrazepam (Fig. 6(b)),

which differs from nitrazepam in having a methyl sub-

stituent on N1 and a fluoro substituent on C20, and

clonazepam, a derivative of nitrazepam that contains a

chloro substituent at the C20 position.

Replacing the C3 hydrogen on nordiazepam with a

hydroxyl group produces oxazepam, an additional substi-

tution of Cl at the C20 position results in lorazepam.

Similarly, replacing the C3 hydrogen on diazepam with a

hydroxyl group produces temazepam. TheCID spectra of the

[MþH]þ ions of these OH-bearing derivatives are dominated

by two common neutral losses: that of water (18Da),

followed by CO (28Da). The spectrum for protonated

oxazepam is shown in Fig. 7. An additional prominent

loss involves the loss of a 56Da neutral. This loss is also

Figure 7. Product ion spectrum of protonated oxazepam,

m/z 287, at Elab¼ 35 eV.


DOI: 10.1002/rcm

Scheme 5.


evident in protonated lorazepam, albeit much less notably.

The proposed mechanism for the loss of water, shown in

Scheme 5, is in agreement with H/D exchange results.31

Nucleophilic attack by the hydroxyl oxygen on the N4

hydrogen, followed by ring contraction and elimination of

water, results in ion XIII at m/z 269. Ring contraction and

cleavage of CO produces ion XIV atm/z 241 that is resonance

stabilized. The 56Da neutral loss directly from protonated

oxazepam is postulated to be that of ethylenedione, C2O2, a

short-lived molecule that rapidly dissociates into two

molecules of CO.42 This hypothesis of C2O2 is consistent

with the absence of the stepwise elimination of the CO

molecules.

CONCLUSIONS

Substitution of chloro, hydroxyl, and nitro groups on the

base diazepam has a dramatic effect on gas-phase dis-

sociation chemistry. The losses of these groups are the most

facile and dominate the CID chemistry. The loss of CO is a

common, facile dissociation pathway of all benzodiazepines,

in evidence even in the aforementioned substituted benzo-

diazepines. Other common losses include the phenyl ring at

the C5 position, as either (substituted) benzene or benzoni-

trile. Ion structures and fragmentation mechanisms have

been postulated towards building a knowledge- and/or

intuition-based benzodiazepine information bank whose

contents will lead to eventual understanding of how

structure drives fragmentation chemistry and determines

products.


AcknowledgementsThis work was supported by the Natural Sciences and

Engineering Research Council (NSERC) of Canada and

MDS SCIEX. A.R. acknowledges financial support from

the University of Calabria to allow her to carry out part of

her thesis research at York University.

REFERENCES

1. Sternbach LHT. J. Med. Chem. 1978; 22: 1.2. Woods JH, Winger G. Psychopharmacology 1995; 118: 107.3. Stewart SH, Westra HA. Current Pharm. Design 2002; 8: 1.4. Jacobsen E. Psychopharmacology 1986; 89: 138.5. Unseld E, Klotz U. Pharm. Res. 1989; 6: 1.6. Olsen RW. Ann. Rev. Pharmacol. Toxicol. 1982; 22: 245.7. Squires RJ, Braestrup C. Nature 1977; 266: 732.8. Mohler H, Okada I. Science 1977; 198: 849.9. Schofield PR, Darlison MG, Fujita N, Burt DR, Stephenson

FA, Rodriguez H, Rhee LM, Ramachandran J, Reale V,Glencorse TA, Seeburg PH, Barnard EA. Nature 1987; 328:221.

10. Sheehan MF, Sheehan DV, Torres A, Coppola A, Francis F.Am. J. Drug Alcohol Abuse 1991; 17: 457.

11. Dussy FE, Hamberg C, Briellmann TA. Int. J. Legal Med.2006; 120: 323.

12. Drummer OH. J. Chromatogr. B 1998; 713: 201.13. Ariffin MM, Miller EI, Cormack PAG, Anderson RA. Anal.

Chem. 2007; 79: 256.14. Verwey B, Eling P, Wientjes H, Zitman FG. J. Clin. Psych.

2000; 61: 456.15. Serfaty M, Masterton G. Br. J. Psychiatry 1993; 163: 386.16. Fitzgerald RL, Rexin DA, Herold DA. Clin. Chem. 1994; 40:

373.17. Kratzsch C, Tenberken O, Peters FT,Weber AA, Kramer T,

Maurer HH. J. Mass Spectrom. 2004; 39: 856.


DOI: 10.1002/rcm


18. Pirnay S, Ricordel I, Libong D, Bouchonnet S. J. Chroma-togr. A 2002; 954: 235.

19. Bugey A, Rudaz S, Staub C. J. Chromatogr. B 2006; 832: 249.20. Quintela O, Cruz A, Concheiro M, de Castro A, Lopez-

Rivadulla M. Rapid Commun. Mass Spectrom. 2004; 18: 2976.21. Zeng H, Deng Y, Wu J. J. Chromatogr. B 2003; 788: 331.22. LeBeau MA, Montgomery MA, Wagner JR, Miller ML.

J. Forensic Sci. 2002; 45: 1133.23. Rivier L. Anal. Chim. Acta 2003; 492: 69.24. Xu J, Huang X. Rapid Commun. Mass Spectrom. 2004; 18: 859.25. Ansari FL, Hussain L, Nasir S, Sultana S. Rapid Commun.

Mass Spectrom. 2005; 19: 1200.26. Gunnar T, Ariniemi K, Lillsunde P. J. Chromatogr. B 2005;

818: 175.27. Zayed MA, Fahmey MA, Hawash MF. Spectrochim. Acta A

2005; 61: 799.28. Dresen S, Kempf J, Weinnmann W. Forensic. Sci. Int. 2006;

161: 86.29. Lips AGAM, Lameijer W, Fokkens RH, Nibbering NMM.

J. Chromatogr. B 2001; 759: 191.30. Weinnmann W, Wiedemann A, Eppinger B, Renz M,

Svoboda M. J. Am. Soc. Mass Spectrom. 1999; 10: 1028.31. Bourcier S, Hoppilliard Y, Kargar-Grisel T, Pechine JM,

Perez F. Eur. J. Mass Spectrom. 2001; 7: 359.32. McCarley TD, Brodbelt J. Anal. Chem. 1993; 65: 2380.33. Smyth WF, McClean S, Ramachandran VN. Rapid Commun.

Mass Spectrom. 2000; 14: 2061.34. Smyth WF, Joyce C, Ramachandran VN, O’Kane E, Coulter

D. Anal. Chim. Acta 2004; 506: 302.


35. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, RobbMA, Cheeseman JR,Montgomery JA, Vreven T, Kudin KN,Burant JC, Millam JM, Iyengar SS, Tomasi J, Barone V,Mennucci B, Cossi M, Scalmani G, Rega N, Petersson GA,Nakatsuji H, Hada M, Ehara M, Toyota K, Fukuda R,Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O,Nakai H, Klene M, X Li, Knox JE, Hratchian HP, Cross JB,Bakken V, Adamo C, Jaramillo J, Gomperts R, StratmannRE, Yazyev O, Austin AJ, Cammi R, Pomelli C, OchterskiJW, Ayala PY, Morokuma K, Voth GA, Salvador P,Dannenberg JJ, Zakrzewski VG, Dapprich S, DanielsAD, Strain MC, Farkas O, Malick DK, Rabuck AD,Raghavachari K, Foresman JB, Ortiz JV, Cui Q, BaboulAG, Clifford S, Cioslowski J, Stefanov BB, Liu G, LiashenkoA, Piskorz P, Komaromi I, Martin RL, Fox DJ, Keith T,Al-Laham MA, Peng CY, Nanayakkara A, Challacombe M,Gill PMW, Johnson B, Chen W, Wong MW, Gonzalez C,Pople JA. Gaussian 03, revision D.01, Gaussian, Inc.: Wall-ingford, CT, 2004.

36. Becke AD. J. Chem. Phys. 1993; 98: 5648.37. Lee CT, Yang WT, Parr RG. Phys. Rev. B 1988; 37: 785.38. Gonzalez C, Schlegel HB. J. Chem. Phys. 1989; 90: 2154.39. Gonzalez C, Schlegel HB. J. Phys. Chem. 1990; 94: 5523.40. Rodriquez CF, Cunje A, Shoeib T, Chu IK, Hopkinson AC,

Siu KWM. J. Am. Chem. Soc. 2001; 123: 3006.41. NIST Chemistry WebBook. Available: http://webbook.

nist.gov/chemistry/.42. Schroder D, Heinemann C, Schwarz H, Harvey JN, Dua S,

Blanksby SJ, Bowie JH. Chem. Eur. J. 1998; 4: 2550.


DOI: 10.1002/rcm

gas-phase fragmentation of protonated benzodiazepines

Documents