Active Hydrogen by Chemical Ionization Mass Spectrometry Yong Yeng Lin and Leland L. Smith Division of Biochemistry, Department of Human Biological Chemistry and Genetics, University of Texas Medical Branch, Galveston, Texas 77550, USA
A means of determination of active hydrogen of hydroxyl, carboxyl, sulfhydryl, amino, amido and sulfonamido groups by chemical ionization mass spectrometry using ammonia and trideuterioammonia as reagent gases is described. The method is based on exchange of active hydrogen for deuterium occurring during the chemical ionization process using trideuterioammonia, with comparison of rn/z for ammonium adduct [M + N&]+ and trideuterioammonium adduct [M- nH + nD ND,]+, or of protonated molecular ion [M+ HI' and deuteronated molecular ion [M - nH +nD D]+ yielding the number of active hydrogens. Applications have been made to several classes of biologically important compounds.
+
+
Demonstration of active hydrogen as a useful first step in recognition of functionality and structure of natural products, biosynthetic intermediates and metabolites available, but in very small amounts, depends upon reliable ultramicro methods not presently. fully exploited. Active hydrogen determinations by chemical ionization mass spectrometry using deuterium labeled reagent gases (D20,ND3,CH30D) has been recorded previously, the active hydrogen of alcohols, amines, etc. exchanging with deuterium of the reagent gas under the conditons of ionization by the reagent gas. We have recently demonstrated the ease with which CIMS with NH3 as reagent gas may be used for recognition of functionality and structure in a series of Cz7steroids: and we report here an extension of these findings which affords a reliable C1 mass spectrometric means of determining active hydrogen in organic compounds by using NH3 and ND3 as reagent gases. Both the number of active hydrogens and the nature of the functional group can be identified concurrently in favorable cases.
EXPERIMENTAL Steroids examined were commercial samples from Steraloids, Inc., Wilton, New Hampshire, o r Research Plus Steroid Laboratories Inc., Denville, New Jersey, or were prepared in our laboratories. Other organic compounds were from various sources. All samples were analyzed by appropriate thin-layer and gas chromatography before use, and in those cases where purity was suspect the samples were purified by high performance liquid chromatography using Waters Associates Inc., Milford, Massachusetts equipment with two p -Porasil columns (4 mm x 30 cm) in series, with hexane + isopropyl alcohol (24: 1, v/v) as mobile phase. Mass spectra were obtained with a Finnigan Corporation model 3200 quadrupole mass spectrometer
equipped with both EI and CI capabilities. Samples (0.21.0 pg) in a 1 cm glass capillary were introduced by direct solid probe into the ionization chamber, with gradual heating from ambient temperature to 200 "C to vaporize samples. Ammonia and trideuterioammonia (99% deuterium) from Merck, Sharp and Dohme Canada Ltd, Montreal, PQ were used as reagent gases, with an ion source temperature of approximately 100 "C (uncorrected meter reading 70 "C) and gas pressure between 0.3-0.5 Torr adjusted to maximize the intensity of the m/z 18 [NH4]+ ion. Higher pressures to 0.9 Torr afforded the same results as regards kind of ions produced, but ion intensities fluctuated with reagent gas pressure.
~~~
RESULTS AND DISCUSSION As was found in our prior study of the CI mass spectra of Cz7-steroid using NH3 as reagent gas, four types of ions were generally formed with the compounds of our present study (Table 1).These ions were the ammonium adduct [M+ NH4]+, the protonated molecular ion [M+ HI', the substitution ion [M+ NH3- XI' and elimination ions [M-XI+. Other ions of lower mass involving multiple eliminations and fragmentation also occurred in certain cases. Analogous ions were found using ND3 as reagent gas, but in analytes containing active hydrogen there was an exchange of hydrogen for deuterium of the isotopic reagent gas ND3 as well. Thus, the ammonium adduct using ND3 as reagent gas is [M + ND4]+ for compounds without active hydrogen (cf. 1, 2 of Table l), but is [M - H + D + ND4]+ for those with one active hydrogen (cf. 3,9,16,19,20)and is [M - n H + n D + ND4]+ for the general case of n active hydrogens. A similar argument holds for the protonated molecular ion [M+H]+ where the general case involving deuterium exchange yields the ion [M - nH + nD + D]'. Comparison of these analogous ions yields direct measure of the number of active hydrogens n,as given in
CCC-0306-042X/79/0006-00 15$02.00
@ Heyden & Son Ltd, 1979
BIOMEDICAL MASS SPECTROMETRY, VOL. 6, NO. 1, 1979 15
Y. Y. LIN AND L. L. SMITH
Table 1. Ammonia CI mass spectra of organic compounds bearing active hydrogens No.
Compound Imol. wt)
Active hydrogens Ammonium adduct
1
Cholest-4-en-3-one(384)
0
2
Cholest-5-en-30-01 30-acetate (428) Cholest-5-en-3p-ol(386)
0
3 4
1-OH
(2OS)-Cholest-5-ene30,20-diol (402) (2OR)-Cholest-5-ene3@,20,21-triol (418) 1lp,l7a,21-Trihydroxypregn4-ene-3,20-dione (362) Estra-l,3,5-( 10)-triene3,170-diol (272)
2-OH
8
Estra-l,3,5-(10)-triene-
3-OH
9
3,16a,17&triol (288) 3-Hydroxybenzo[alpyrene (268)
1-OH
10
Morphine (285)
2-OH
11
D-Glucose (180)
5-OH
5 6
7
12
13
14
15
16
17
2-Deoxy-P-amino-~glucose (179)
N-Acetyl-2-deoxy-2amino-D-glucose (221)
3-OH 3-OH 2-OH
4-OH l-NH2
4-OH 1-CONH-
2-Diethylaminoethyl-p1-NH2 aminobenzoate (Procaine) (236) Phenylephrine 1-NH (rn-HOC6H4CHOHCH2NHCH3) 2-OH (167) Phenacetine 1-CONH (p-C2H5OC6H4NHCOCH3) ( 179) p-Acetoxymethylbenzene1-S02NH2 sulfonamide (229) a-Acetoxytol butamide (p-CH3C02CHZCeH4S02 NHCONHCdH,) (328)
1-so2 NHCONH-
1-SH
20
21-Mercaptopregn4-ene-3-20-dione(346) Oleic acid (282)
21
L-Serine (105)
18
19
1-COOH 1-COOH 1-NH2 1-OH
402(100) 406 (100) 446 (100) 450 (100) 404 (100) 409 (100) 420 (5) 426 (9) 436 (100) 443 (100) 380 (18) 387 (8) 290 (30) 296 (41) 306 (100) 313 (100)
-
303 (100) 309 (100) 198 (100)
Protonated molecular ion
Othei ions
385 (78) 386 (96)
-
-
363 (49) 367 (14) 273 (60) 276 (20) 289 (3) 293 (5) 269 (100) 271 (100) 286 (66) 289 (64)
-
207 (100)
-
197 (15)
180 (100)
207 (22)
187 (100)
239 (6)
222 (15)
248 (3)
228 (6)
-
237 (100) 240 [ 100)
197 (100) 202 (1 00)
180 (29) 182 (30)
247 (100) 253 (100)
-
346 (100)
329 (20)
352 (100)
332 (8)
364 (100) 369 (100) 300 (100) 305 (100) 123 (52)
347 (81) 349 (72)
-
-
106 (100)
367 (20Ib 367 (16)b
-
320 (49)'. 303 (1001' 325 (55)", 305 (100)' 272 274 288 (5)d 291 (3)d
-
162 (1)", 138 (5)". 120(1)", 108(l)' 145 (1)". 167 (l)", 112 (3)" 162 (5)". 138 (30)', 120 (10)'. 108 (5)" 167 (7)". 145 (64)", 125(15)",112(34)" 204 (30)", 138 (50le, 119 (70)", 102 209 (56)". 145 (100)'. 125 (64)". 105 (78)"
-
264 (7)', 189 (13) 273 (4Ig, 196 (181, 195 (111 288 (151,247 (20). 134(100), 117(100) 295 (131,294 (8). 253 (151, 141 (96). 121 (75)
140 (31f
-
131 (17) 111 (100) 156 (13)f 139 (100) 122 (83) 1-COOH l-NH2 167 (3)' 1-SH a CI mass spectra obtained using a Finnigan Corporation model 3200 quadrupole mass spectrometer and NH3 or ND3 as reagent gas at 0.5 Torr, with ionization source temperature 100 "C. Spmples (0.1-1 wg) were introduced via the direct inlet solids probe. Only ions above 100 amu not obviously due to isotope abundance and not less than 1% abundance are recorded. [MI' probably due t o charge transfer process, cf. Ref. 5. Double elimination ion [M-H,O-OHl+. f c Ions from sidechain scission. " Multiple elimination ion of aldohexose. [M+N2H71f. * IM- nH+ nD+N,D71'.
22
L-Cysteine (121)
16 BIOMEDICAL MASS SPECTROMETRY, VOL. 6, NO. 1, 1979
@ Heyden & Son Ltd, 1979
ASSESSMENT OF ACTIVE HYDROGEN
Eqn (1): n =( m / z ) N D 3 - ( m / z ) ~ ~ 3 - a
(1)
where a = 4 for the ammonium adduct, a = 1 for the protonated molecular ion. However, for the substitution and elimination ions, special considerations must be made, as the eliminated moiety X may contain active hydrogen. The substitution ion using ND3 as reagent gas is actually [M - n H + nD + ND3-XI', the elimination ion [M-nH+nD-XI+. Where X does not contain active hydrogen (cf. 2), n may be determined using Eqn (1) with a = 3 for the substitution ion, with a = 0 for the elimination ion. Where X contains active hydrogen (cf. 3-5, 11)values of n calculated in this fashion must be increased by one. Compounds 3-22 bearing active hydrogens which undergo essentially complete exchange with deuterium are all analyzed accurately by these data, as are other compounds not reported here. The number of hydroxyl active hydrogens is correctly determined in simple alcohols (3,9),in polyols (4,5, 7, 8) and in multifunctional compounds (6,10-13,15, 21), as are also active hydrogens of carboxyl, amino, amido, sulfonamido, sulfonylurea and thiol compounds, present singly or in combinations in multifunctional derivatives. Where there is no active hydrogen (cf. 1, 2), the spectra confirm no active hydrogen. In distinction to use of D2O as reagent gas,2.5*6 no isotope incorporation from ND3 into unsaturated compounds such as aromatic compounds, ketones, enones, aldehydes, or esters was observed. However, some enolizable 1,3-dicarbonyi systems were found to undergo isotope exchange with ND3. Thus, ND3 CI mass spectra of 2-methylcyclopentane-1,3-dione evinced essentially complete exchange, and spectra of ethyl acetoacetate and malonate showed 30% and 5% exchange respectively (see Fig. 1). In addition to active hydrogen determination, these CI mass-spectra aid in recognition of functionality as well. Aliphatic alcohols (3-5,11) are characterized by prominent ammonium adducts accompanied by substitution and elimination ions but not by the protonated molecular ion.4 Other ionization processes involving other functionality in multifunctional compounds may also occur and give rise to other ions which are more abundant. Thus, in spectra of 6, ions derived from sidechain scission ( m / z 305, 320 with NH3) and the protonated moiecular ion associated with the enone feature are dominant over the ammonium adduct associated with hydroxylic functionality. Phenolic hydroxyls (7-10)are distinguished from aliphatic hydroxyls by the presence of the protonated molecular ion and the absence of both substitution and elimination ions. Interestingly, [MI+' ions are observed in the NH3 CI mass spectra of phenols 7 ( m / z 272) and 8 ( m / z 288). These two ions have the same m / z ratio as NH3 substi-
I61t M t H I + r
1
0
I
I
100
162[M+Dlt 1
I
200
I00
2
m/ z
Figure 1. Chemical ionization mass spectra of enolizable 1,3dicarbonyl compounds using NH,and ND,as reagent gases. Ions due to "Hal+, [ND.$, "2H71+ and [N2D71+have been omitted.
+
tution ions [M NH3 - OH]' for 7 and 8, but are in fact molecular ions [MI" as evinced by ND3 CI mass spectra. With ND3, ions m / z 274 for 7 and m l z 291 for 8 establish that these ions are [M-nH+nD]+ and not [M- nH+nD+ND3-OH]+. These [M]"ionsobserved for the phenols 7 and 8 are not likely to have been formed by E I processes, but may derive by charge transfer processes between analyte and reagent gas. Similar [MI+' ions have been observed in NH3 CI mass spectra of alkyl substituted benzenes7 and in D 2 0 CI mass spectra of toluene5 The higher proton affinityof the amino group relative to NH3 is evinced in NH3 CI spectra of amines (12,14, 15) by intense proton complex ions [M+H]+. Zwiterionic amino groups of the amino acids (21,22) respond similarly, but in addition the amino acid carboxyl is represented by a prominent ammonium adduct ion as well. In amides (13,16) with diminished proton affinities both the proton complex [M+ H]+ and ammonium adduct [M + NH4]+ are prominent, whereas in sulfonamide (17)and sulfonylurea (18)derivatives of yet greater diminished proton affinity the ammonium adduct [M + NH4]+ is the predominant ion. Judicious application of these approaches using ammonia CI mass spectra to recognition of functionality and to structure problems appears warranted. However, proper caution should be exercised in extension of the method to new or unknown compound types.
Acknowledgements Financial support of these studies was provided by the US Public Health Service under research grant HL-10160. Samples of drugs and sulfonamide derivatives were generously donated by Dr R. R. Kernpen and Mr S. F. Micheletti of this university.
REFERENCES 1. D. F. Hunt, C. N. McEwen and R. A. Upham, Tetrahedron L e n 4539 (1971). 2. D. F. Hunt, C. N. McEwen and R. A. Upham. Anal. Chem. 44, 1292 (1972).
@ Heyden & Son Ltd, 1979
3. W. Blum, E. Schlumpf, J. G. Liehr and W. J. Richter, Terrahedron Len. 565 (1976). 4. Y. Y. Lin and L. L. Smith American Society for Mass Spectrometry, 25th Annual Conference on Mass Spectrometry and
BIOMEDICAL MASS SPECTROMETRY, VOL. 6, NO. 1, 1979 17
Y. Y. LIN AND L. L. SMITH Allied Topics, Washington, DC (1977). Abstracts, p. 134 (pp. 533-534 of extended abstracts). 5. D. P. Martinesen and S.E. Buttrill, Org. MassSpectrom. 11,762 (1976). 6. B. S.Freiser, R. L. Woodin and J. L. Beauchamp, J. Am. Chem. SOC.97, 6893 (1975). 7. A. Tatematsu, M. Suzuki, Yoshizumi, K. Harada and H.Nakata, American Society for Mass Spectrometry 25th Annual Con-
18 BIOMEDICAL MASS SPECTROMETRY, VOL. 6, NO. 1, 1979
ference on Mass Spectrometry and Allied Topics, Washington, DC (1977).Abstracts, p. 63.
Received 9 June 1978
@ Heyden & Son Ltd, 1979
@ Heyden & Son Ltd, 1979