A simple criterion for gas chromatography/mass spectrometric analysis of thermally unstable compounds, and reassessment of the by-products of alkyl diazoacetate synthesis

Article abstract of DOI:10.1002/rcm.6457

Rationale: A principal limitation of gas chromatography (GC) and gas chromatography/mass spectrometry (GC/MS) is the thermal instability of analytes. We propose that the injector and column temperatures should not exceed the atmospheric pressure boiling point, without decomposition, of the highest homologue of the series being analyzed, instead of the time-consuming procedure of obtaining chromatograms using different temperatures. Methods: A series of thermally unstable diazocarbonyl compounds, alkyl diazoacetates (predicted limit of stability approx. 140 °C, the boiling point of ethyl diazoacetate), was selected for GC/MS analysis using standard equipment. Different GC separation conditions were selected so that the retention temperatures of target compounds were both below and above 140°C. Results: Analyzing alkyl diazoacetates within their thermal stability range permitted reanalysis of their typical synthesis by-products. No dialkyl fumarate or maleate impurities, principal decomposition products which have often been reported previously, were found. Instead, alkyl esters of glycolic acid nitrate, O₂NOCH ₂CO₂R, and 'pseudo-dimeric' products, ROCO[C ₂H₃NO]CO₂R, were discovered for the first time. Conclusions: Avoiding the decomposition of thermally unstable organic compounds during GC and/or GC/MS analysis requires estimating their degradation temperature limits. This limit can be estimated as being equal to the atmospheric pressure boiling point of the highest homologue in the homologous series under consideration that does not decompose on boiling. Copyright

Full text of DOI:10.1002/rcm.6457

Research Article
Received: 12 September 2012
Revised: 25 October 2012
Accepted: 26 October 2012
Published online in Wiley Online Library
Rapid Commun. Mass Spectrom. 2013, 27, 461–466
(wileyonlinelibrary.com) DOI: 10.1002/rcm.6457
A simple criterion for gas chromatography/mass spectrometric
analysis of thermally unstable compounds, and reassessment of
the by-products of alkyl diazoacetate synthesis
*
Tatiana A. Kornilova, Anton I. Ukolov, Rafael R. Kostikov and Igor G. Zenkevich
Department of Chemistry, St. Petersburg State University, Universitetskii prosp. 26, St. Petersburg 198504, Russia
RATIONALE: A principal limitation of gas chromatography (GC) and gas chromatography/mass spectrometry (GC/MS) is
the thermal instability of analytes. We propose that the injector and column temperatures should not exceed the atmospheric
pressure boiling point, without decomposition, of the highest homologue of the series being analyzed, instead of the
time-consuming procedure of obtaining chromatograms using different temperatures.
METHODS: ꢀA series of thermally unstable diazocarbonyl compounds, alkyl diazoacetates (predicted limit of stability
approx. 140 C, the boiling point of ethyl diazoacetate), was selected for GC/MS analysis using standard equipment.
Different GC separation conditions were selected so that the retention temperatures of target compounds were both
ꢀ
below and above 140 C.
RESULTS: Analyzing alkyl diazoacetates within their thermal stability range permitted reanalysis of their typical synthesis
by-products. No dialkyl fumarate or maleate impurities, principal decomposition products which have often been reported
previously, were found. Instead, alkyl esters of glycolic acid nitrate, O2NOCH2CO2R, and ’pseudo-dimeric’ products, ROCO
[C2H3NO]CO2R, were discovered for the first time.
CONCLUSIONS: Avoiding the decomposition of thermally unstable organic compounds during GC and/or GC/MS
analysis requires estimating their degradation temperature limits. This limit can be estimated as being equal to the
atmospheric pressure boiling point of the highest homologue in the homologous series under consideration that does
not decompose on boiling. Copyright © 2012 John Wiley & Sons, Ltd.
One of the principal limitations of gas chromatography (GC)
and gas chromatography/mass spectrometry (GC/MS) tech-
niques is the thermal instability of analytes. This limitation
can be avoided if the injector and chromatographic column
temperatures do not exceed certain limits, which should be
estimated experimentally, or predicted, prior to analysis.
Conventional practice with contemporary GC equipment is
adequate for the analysis of such unstable compounds as
organic peroxides,[1] but there are no general criteria for
evaluating the applicability of GC/MS instruments to very
different thermally labile compounds. Analytical methods that
exclude the heating of samples containing such compounds are
preferable at the moment. Thus, the latest recommendations for
the analysis of traces of ethyl diazoacetate are to use high-
performance liquid chromatography (HPLC) rather than GC.[2]
Thermal lability also underlies the preference for mass
spectrometry as the analytical method for various diazo
compounds (RR’C = N+ = Nꢁ). Mass spectra of several dozen
diazo compounds are presented in the NIST/NIH/EPA MS
database.[3] Special publications are devoted to the discussion
of MS properties suitable for diazomethane,[4] some alkyl
diazoacetates, methyl diazoacetoacetate, dimethyl diazoma-
lonate, 3-diazopentan-2,4-dione,[5] and others. The minimal
temperature restrictions for the registration of electron ioniza-
tion (EI) mass spectra also make this method preferable for
the analysis of complex polyfunctional diazo compounds.[6,7]
The main GC characteristics of analytes are their retention
indices (RI),[8] primarily on standard non-polar stationary
phases. However, experimental RI values remain unknown
even for the simplest diazo compounds. RI evaluations based
on the additive scheme,[9] which are included in the NIST MS
database[3] for a great number of compounds, appear to be
unavailable for diazo compounds because of the ’unusual’
valence state of nitrogen atoms in these molecules.
This paper is devoted to evaluating the thermal stability
limits of diazocarbonyl compounds, primarily alkyl diazoace-
tates (I), for the purpose of their GC/MS analysis. To achieve
this it is necessary to reconsider the set of by-products present
in the reaction mixtures after these esters are synthesized.
* Correspondence to: I. G. Zenkevich, Department of Chemistry,
St. Petersburg State University, Universitetskii prosp. 26,
St. Petersburg 198504, Russia.
R ¼ MeðaÞ; EtðbÞ; Pr ðcÞ; i-Pr ðdÞ; BuðeÞ;
i-Bu ðfÞ; t-BuðgÞ:
E-mail: izenkevich@mail15.com
Rapid Commun. Mass Spectrom. 2013, 27, 461–466
Copyright © 2012 John Wiley & Sons, Ltd.

T. A. Kornilova et al.
Another compound containing a similar diazo fragment,
namely 1-diazo-4-phenylbutan-2-one (II), was synthesized
to compare its thermal stability with I.
0.25 mm, film thickness 0.25 mm). The temperature program
–1
was: 40 ꢀC, 5 C min to 250 C. The carrier gas was helium,
with a flow rate of 2 mL min–1, and the reagent gas was
methane, with a flow rate of 0.4 mL minꢁ1. High-resolution
measurements were carried out using an Agilent 6250 QTOF
LC/MS system equipped with a reversed-phase SL column,
an eluent of 90% water containing 20 mM ammonium formate
and 10% acetonitrile containing 0.05% triethylamine, at a flow
rate of 0.15 mL min–1.
ꢀ
ꢀ
EXPERIMENTAL
Alkyl diazoacetates (I) were synthesized from hydrochlorides
of the corresponding aminoacetic acid alkyl esters and
sodium nitrite in acid media, in accordance with the general
method published previously,[10] using dichloromethane or
diethyl ether as the solvent:
The retention temperatures (TR) of the alkyl diazoacetates
were evaluated using the following relationship:
À
Á
TR ¼ T0 þ r tR ꢁ tTð0Þ
(1)
where Т0 is the initial temperature (ꢀС), r is the ramp (ꢀС minꢁ1),
tR is the retention time of the analyte (min), and tT(0) is the
duration of the initial isothermal step (min).
ꢁ
ROCOCH2NH3þ þ NO2 ! ROCOCHN2 þ 2H2O
1-Diazo-4-phenylbutan-2-one (II) was synthesized from
diazomethane and 3-phenylpropionyl chloride according to
published recommendations:[11]
Retention times of the C8–C20 n-alkanes, which were
required to calculate the retention indices, were measured
under the same GC conditions in parallel runs.
C6H5ðCH2Þ2COCl þ CH2N2 ! C6H5ðCH2Þ2COCHN2 þ HCl
Most analyses of alkyl diazoacetates were carried out using
a Shimadzu QP 2010plus GC/MS system equipped with a
WCOT column with HP-5 stationary phase (column length
25 m, i.d. 0.20 mm, film thickness 0.33 mm). A ’smooth’
temperature program (A) was used: 45 ꢀC (1 min),
5 ꢀC min–1 to 280 ꢀC, then held for 2 min. The helium carrier
gas linear flow was 40 cm s–1, and a split ratio of 1:200 and an
RESULTS AND DISCUSSION
The analysis of alkyl diazoacetates (I) using temperature
program A allowed their separation without thermal decompo-
sition. The mass spectra of esters Ia–g are presented in Table 1,
and all of the spectra are in complete accordance with known
regular MS fragmentation patterns.[12] The molecular ions (M,
indicated in bold) were found in all spectra, even the spectra
of Ig (R = t-Bu), and are typical of compounds with conjugated
systems caused by charge delocalization and a molecular ion
containing an unpaired electron (CO-CH = N+ = N– for
diazocarbonyl compounds). The most characteristic ions for
the alkyl diazoacetates were [М – 28] = [M – N2]+, [M – RO]+ =
[N2CHCO]+ (m/z 69), [M – N2 – RO]+ = [CHCO]+ (m/z 41),
and, if R = Pr, [N2CHC(OH)2]+ (m/z 87, the result of ’double
ꢀ
injector temperature of 100 C were used. Other instruments
and various, more ’harsh’, temperature regimes were used
for comparison and for modeling the thermal decomposition
of the diazo compounds (as indicated below).
To determine the molecular weight of partially identified
by-products, VI, the compound VIb (R= Et) was characterized
using a positive chemical ionization (PCI) mass spectrum
obtained on an Agilent 7000 GC-QQQ GC/MS system
equipped with a WCOT column (HP-5, length 30 m, i.d.
Table 1. Retention temperatures (TR), retention indices (RI), and mass spectra (70 eV) of alkyl diazoacetates, N2CHCO2R (I)
R, №, MW
Me, Ia, 100
Et, Ib, 114
Pr, Ic, 128
TR ꢀC
54
RI
Mass spectrum, m/z ≥40 (Irel ≥2%)
784 ꢂ 5
867 ꢂ 2
959 ꢂ 2
101(4), 100(100)M, 72(2) [M – N2], 70(2), 69(78) [N2CHCO], 59(4), 44(3),
43(23), 42(10), 41(51) [CHCO], 40(3).
115(5), 114(82)M, 86(9) [M – N2], 70(3), 69(100) [N2CHCO], 68(2), 58
(15), 57(9), 45(8), 44(3), 43(11), 42(19), 41(39) [CHCO], 40(2).
128(14)M, 100(2) [M – N2], 99(2), 87(24) [N2CHC(OH)2], 86(5), 70(2), 69
(58) [N2CHCO], 59(4), 58(6), 57(3), 56(3), 55(5), 54(2), 44(4), 43(100)
[C3H7], 42(18), 41(80) [CHCO], 40(3).
128(19)M, 87(3), 86(3), 69(47) [N2CHCO], 59(5), 55(2), 45(2), 44(4), 43
(100) [C3H7], 42(11), 41(44) [CHCO], 40(2).
142(3)M, 114(4) [M – N2], 113(2), 101(6), 99(3), 87(26) [N2CHC(OH)2],
86(4), 73(3), 72(2), 71(2), 70(2), 69(47) [N2CHCO], 68(2), 59(2), 58(5),
57(51) [C4H9], 56(34), 55(22), 54(2), 53(2), 44(4), 43(12), 42(16), 41(100)
[CHCO], 40(3).
62
73
i-Pr, Id, 128
Bu, Ie, 142
66
87
902 ꢂ 1
1058 ꢂ 1
i-Bu, If, 142
t-Bu, Ig, 142
81
71
1015 ꢂ 1
940 ꢂ 2
142(3)M, 99(5), 88(2), 87(58) [N2CHC(OH)2], 86(2), 73(2), 71(2), 70(4),
69(100) [N2CHCO], 59(5), 57(62) [C4H9], 56(40), 55(10), 53(2), 44(2), 43
(25), 42(13), 41(95) [CHCO], 40(3).
142(0.3)М, 70(5), 69(68) [N2CHCO], 68(2), 67(5), 57(4), 54(3), 53(10), 51
(3), 43(5), 42(5), 41(100) [CHCO], 40(3).
wileyonlinelibrary.com/journal/rcm
Copyright © 2012 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2013, 27, 461–466

GC/MS analysis of thermally unstable compounds
hydrogen rearrangement’[12]). However, if GC separation of the
diazoesters I was ꢀperformed under more ’harsh’ conditions
ROCOCH = CHCO2R, formed as a result of decomposi-
tion in condensed media followed by intermolecular
dimerization.[14]
(regime B, T0 = 70 C, r = 10 C minꢁ1), no diazo compounds
ꢀ
were found in the chromatograms, but peaks indicating
products of stationary phase degradation appeared. This fact
confirmed the existence of a thermal stability limit for the GC
and GC/MS analysis of diazocarbonyl compounds, which
needed to be evaluated properly.
It is important to note that such by-products (for most of
which mass spectra and RI data are known[3]) are not present
in diazoesters I, meaning that their decomposition before
analysis is completely suppressed. However, diazoester I sam-
ples do contain some unidentified constituents (an observation
that has been mentioned previously[2]). The consistency of the
relative chromatographic peak areas under different GC/MS
conditions confirms the formation of such products before
analyses (i.e., during synthesis).
A stability criterion for thermally unstable compounds
during GC and/or GC/MS analysis
The simplest homologues from a series of thermally unstable
compounds can often be distilled at atmospheric pressure
without their thermal decomposition and, therefore, can be
characterized by ’normal’ boiling points at atmospheric
pressure (Tb, ꢀC). Beyond a certain homologue, heating higher
members of the series to boiling point becomes impossible
because of their decomposition. Therefore, we can select the
atmospheric pressure boiling point of the highest homologue
of the series under consideration, Tmax, as the GC and/or
GC/MS analysis temperature limit for that homologous series.
If the chromatographic column and/or injector temperature
(or the ’retention temperature’, TR [see Eqn. (1)], if a tempera-
ture program is used) exceeds Tmax, GC analysis of the
compounds becomes impossible. Such a criterion appears to
be the simplest method of estimating the maximum analysis
temperature for both GC and GC/MS. The alternative method
of evaluating the thermal stability of analytes is to carry out
time-consuming experiments at various injector and/or
chromatographic column temperatures to determine the limit
of their thermal stability.
The aliphatic azides series RN3 is used as an example here.
Tb values for the series are (in ꢀC): 20–21 (R = Me), 48–50 (R = Et),
77.5–78 (R= Pr), 106.5 (R= Bu), and 130–135 (R = n-C5H11).
Homologues with R = C6H13 have no boiling points at
atmospheric pressure. Therefore, GC/MS analysis of these
compounds should be possible up to retention temperatures
of approx. 130–135 ꢀC, which has been confirmed experimen-
tally.[13] Within the alkyl hypochlorites series, ROCl, Tb values
are known only for homologues with R = Pr (62.9 ꢀC) and
R = t-Bu (77–78 ꢀC), meaning that Tmax is approx. 80 ꢀC, which is
high enough for RI values to be determined for ethyl (RI= 502)
and tert-butyl hypochlorites (RI= 605). Note that RI values
determined using only standard non-polar polydimethyl
siloxane stationary phases are discussed throughout this paper.
Ethyl diazoacetate (Ib) is one of the most often used aliphatic
diazocarbonyl reagents, and the physicochemical properties
available for this compound (including Tb 140–143 ꢀC) appear
to be the most reliable for its homologous series. We could not
find any Tb values for higher alkyl diazoacetate homologues
in the literature. Therefore, GC and/or GC/MS analysis of dia-
zocarbonyl compounds that possess the fragment -CO-CHN2
should be possible up to a temperature of approx. 140 ꢀC.
Generally, for the GC and/or GC-MS analysis of thermally
unstable compounds, three kinds of decomposition should be
distinguished:
(ii) Thermal decomposition or isomerization of analytes within
the heated injector of the chromatographic system. These
processes can be revealed by changing the injection
temperature, resulting in changes in the relative peak
ꢀ
areas. An injector temperature of 100 C was selected for
the analysis of alkyl diazoacetates because it avoids ther-
mal degradation of the analytes. An injection temperature
of 77 ꢀC, which has previously been recommended,[2]
appeared to be too low, leading to broadening of the
chromatographic peaks.
(iii) Thermal decomposition of the analytes within the chromato-
graphic column leads to the disappearance of the chromato-
graphic peaks or the appearance of broad diffuse signals
between the initial compound and product peaks. This
mode of decomposition is illustrated by the chromatogram
(Fig. 1) of 1-diazo-4-phenylbutan-2-one (II) under GC
Figure 1. Fragment of the chromatogram of 1-diazo-4-
phenylbutan-2-one (conditions C, TR 145 ꢀС): the peak at reten-
tion time 46.3 min is 4-phenyl-1-chlorobutan-2-one, the main
peak (51.5–52.3 min) is 1-diazo-4-phenylbutan-2-one, and the
peak at 58.7 min is an unidentified component. The ’train’ in
front of the main peak resulted from the partial decomposition
of the diazo compound in the GC column because its retention
temperature exceeded the Tmax value.
(i) Decomposition of target compounds prior to analysis
(i.e., during their synthesis, isolation, or storage). Typical
degradation products for alkyl diazoacetates are dialkyl esters
of (E)- and (Z)-but-2-en-dioic acids (fumarates and maleates),
Rapid Commun. Mass Spectrom. 2013, 27, 461–466
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

T. A. Kornilova et al.
conditions C (T0 =40 ꢀC, r = 2 ꢀC minꢁ1) when its TR value
(~145 ꢀC) slightly exceeds the Tmax estimated for diazocar-
bonyl compounds.
The differences between the RIs of alkyl chloroacetates and
diazoacetates were fairly constant, with an average value of
ꢁ25 ꢂ 3 i.u.
Most impurities with retention times higher than those of I
were characterized by unique mass spectra, which were not
able to be identified using the NIST database.[3] Not only these
specific compounds, but their entire classes, are absent from
this database. The first impurities were also characterized by
constant ΔRI values relative to the alkyl diazoacetates (average
ΔRI +112ꢂ 8) and their mass spectra included intense peaks for
the [R]+ ion, and at m/z 46 and 76 (see Table 3). The m/z 46
([NO2]+) and 76 ([CH2ONO2]+) peaks are specific to aliphatic
esters of nitric acid.[12] This combination of MS and GC
analytical parameters allowed us to identify the impurities as
alkyl esters of glycolic acid nitrate, O2NOCH2CO2R (IV). The
concentrations of these impurities in the reaction mixtures
varied widely and were comparable with those of I, as
illustrated in Fig. 2.
The ’train’ in front of the main peak is a result of the partial
decomposition of the diazocarbonyl compound, with loss of
nitrogen and the formation of 4-phenylbut-1-en-1-one (a substi-
tuted ketene), which was confirmed by its mass spectrum.
As one can see from the retention temperature values for
alkyl diazoacetates (I) (Table 1), their retention temperatures
under conditions A are 54–87 ꢀC, much lower than their Tmax
(~140 ꢀC), meaning that all of these compounds can be
considered stable during GC/MS analysis under these
conditions. Any attempt to change the analytical conditions
so that TR >Tmax leads to the complete disappearance of the
diazo ester peaks. Control of the TR values is strongly
recommended for the GC/MS analysis of other thermally
unstable compounds.
The formation of nitric acid esters can be ascribed both to
the presence of uncontrolled amounts of sodium nitrate in
the sodium nitrite used in the synthesis and to more complex
diazo compound reactions. For example, it is known that the
Reassessment of the by-products of alkyl diazoacetates
formed in the reaction mixtures
Having proved the thermal stability of alkyl diazoacetates
during their GC/MS analysis using the criterion TR <Tmax
we were able to reassess the typical by-products formed in
alkyl diazoacetate reaction mixtures.
All alkyl diazoacetates contain corresponding chloroacetate
impurities (IIIa–f) resulting from the reaction of diazo esters
with the hydrogen chloride used in their synthesis:
ROCOCHN2 þ HCl ! ROCOCH2Cl þ N2
All of the compounds (IIIa–f) could be unambiguously
identified using known mass spectra. Moreover, the validity
of their identification was confirmed using RI values (Table 2).
Table 2. Retention indices (RI) of alkyl chloroacetates,
ClCH2CO2R (III)
R, №
RI
R (№)
RI
Figure 2. Chromatogram of the isobutyl diazoacetate reaction
mixture (Ie): the peak at retention time 7.41 min is isobutyl
chloroacetate (IIIe), the peak at 8.19 min is isobutyl diazoace-
tate, and the peak at 11.18 min is the isobutyl ester of glycolic
acid nitrate (IVe), O2NOCH2CO2CH2CH(CH3)2.
Me, IIIa
Et, IIIb
Pr, IIIc
763 ꢂ 2
839 ꢂ 2
933 ꢂ 1
i-Pr (IIId)
Bu (IIIe)
i-Bu (IIIf)
880 ꢂ 1
1032 ꢂ 2
988 ꢂ 1
Table 3. Retention indices (RI) and mass spectra of the by-products present in alkyl diazoacetates: alkyl esters of glycolic acid
nitrate, O2NOCH2CO2R (IV)
R, №
RI
Mass spectrum, m/z ≥40 (Irel ≥2%)
Me, Va
Et, Vb
i-Pr, Vd
i-Bu, Vf
908 ꢂ 2
977 ꢂ 2
1010 ꢂ 2
1120 ꢂ 2
135(0.6)М, 77(4), 76(38) [O2NOCH2], 60(3), 59(92) [CO2CH3], 46(100) [NO2], 45(23), 44
(6), 43(2), 42(4).
149(0.1)М, 106(2), 90(4), 76(51) [O2NOCH2], 61(2), 60(2), 59(2), 57(4), 46(100) [NO2], 45
(15), 44(8), 43(10), 42(6).
163(0)M, 148(2), 104(2), 90(25), 76(13) [O2NOCH2], 59(2), 46(56) [NO2], 45(7), 44(5), 43
(100) [C3H7], 42(5), 41(24).
177(0)M, 116(1), 104(2), 76(32) [O2NOCH2], 61(7), 59(3), 58(6), 57(100) [C4H9], 56(51), 55
(7), 46(53) [NO2], 44(5), 43(50), 42(11), 41(74), 40(2).
wileyonlinelibrary.com/journal/rcm
Copyright © 2012 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2013, 27, 461–466

GC/MS analysis of thermally unstable compounds
Table 4. Retention indices (RI) and mass spectra of the by-products present in alkyl diazoacetates: ’pseudo-dimeric’ constituents,
[C2H3NO](CO2R)2 (VI)
R, №
RI
Mass spectrum, m/z ≥40 (Irel ≥2%)
Et, VIb
Pr, VIc
1460 ꢂ 3
1696 ꢂ 3
158(4), 131(5), 130(57), 103(10), 102(100), 77(6), 75(7), 74(40), 60(3), 58(4), 57(4), 56(19), 47
(3), 45(6), 44(44), 43(5), 42(6), 41(2).
172(2), 159(14), 141(2), 130(11), 100(2), 97(2), 84(2), 73(2), 70(2), 59(4), 57(2), 55(3), 54(12),
53(5), 44(15), 43(100), 42(37), 41(33), 40(2).
i-Pr, VId
Bu, VIe
1586 ꢂ 4
1869 ꢂ 5
172(4), 157(3), 155(2), 130(16), 100(2), 69(2), 59(3), 53(3), 45(7), 44(5), 43(100), 42(3), 41(16).
186(1), 159(14), 130(6), 98(2), 73(4), 70(2), 58(4), 57(100), 56(71), 55(16), 54(10), 53(5), 52(2),
45(2), 44(15), 43(24), 42(7), 41(75), 40(3).
t-Bu, VIg
1573 ꢂ 4
130(4), 58(4), 57(100), 56(6), 55(3), 53(3), 43(7), 42(2), 41(29).
interaction of R’CHN2 with N2O5 and, possibly, with other
nitrogen oxides, gives nitric esters R’CH2ONO2 as well as
nitro derivatives R’C(NO2)N2.[15]
244 [M + C3H5]+, which are typical for a PCI mass spectrum
using methane. The relative intensity of the isotopic m/z 205
peak was 9.6% (the calculated value for C8H14NO5 is 9.4%).
High-resolution measurements on the [M + H]+ ion (m/z 204)
were performed using an Agilent 6250 QTOF LC/MS system,
and gave an m/z value of 204.0887 (the calculated mass for
C8H14NO5 is 204.0872 Da). Therefore, we propose that this
reaction product, which has an MW of 203, has the structure
C2H5OCO[C2H3NO]CO2C2H5, although the origin of the
C2H3NO fragment remains unknown and needs to be revealed
in future work. The general formula for compounds VI can,
therefore, be written as ROCO[C2H3NO]CO2R.
The formation of the corresponding nitrous acid ester,
ONOCH2CO2R (V), was observed in only one case, when
R = Et (RI 904 ꢂ 2, (m/z)100 = 60 [ONOCH2]). The reaction mix-
ture for the synthesis of ester Ib contained small amounts of
the ethyl ester of hydroxyacetic acid (ethyl glycolate),
HOCH2CO2C2H5, which was identified by comparing its
mass spectrum with the NIST database,[3] and its RI value
(RI 783 ꢂ 2, RIref 788 ꢂ 15; mass spectrum 104(-)M, 77(4), 76
(23) [M – C2H4], 75(7), 74(2), 73(3), 62(2), 61(71), 60(3), 59(37)
[HOCH2CO], 58(5), 47(5), 46(4), 45(100) [C2H5O], 44(7), 43
(21), 42(30), 41(3)), and comparing its mass spectrum with
other compounds with mass spectra that can be interpreted
as being for the ethyl ester of glioxylic acid oxime
C2H5OCOCH = NOH (RI 1062 ꢂ 1; mass spectrum 117(-)M,
100(21) [M ꢁ OH], 90(4), 89(100) [M – C2H4], 86(4), 84(5), 74
(4), 73(17), 72(68) [M ꢁ C2H4 ꢁ OH], 71(50), 58(4), 57(3), 56
(5), 55(5), 54(17), 46(2), 45(26), 44(45), 43(16), 42(3), 41(2)).
Apart from the impurities mentioned above, the alkyl
diazoacetate reaction mixtures also contained more ’unusual’
constituents (VI), the mass spectra of which are presented
in Table 4.
As discussed above, by-products are formed during the
reaction of amino acetates with nitrous acid, but nitroglycolic
acid esters and compounds VI are absent from 1-diazo-4-
phenylbutan-2-one because it is synthesized in a different way.
CONCLUSIONS
Avoiding the decomposition of thermally unstable organic
compounds during GC and/or GC/MS analysis requires
stable temperature limits to be estimated. A simple way of
estimating these limits is to use the boiling points, without
decomposition, of the highest homologues of the series under
consideration, at atmospheric pressure. For diazocarbonyl
compounds the limit is approx. 140 ꢀC.
No molecular ion signals were observed for these constitu-
ents, and the most intense peaks confirmed only the presence
of alkyl fragments, R. All of the mass spectra contained an
m/z 130 peak, which is typical of dialkyl esters of dicarboxylic
Using stable GC/MS analysis of alkyl diazoacetates
allowed us to reassess the typical by-products formed during
their synthesis by diazotation of aminoacetic acid alkyl esters.
acids, X(CO2CnH2n+1)2, and corresponds to the consecutive
[12]
elimination of the fragments CnH2n+1O and CnH2n
(i.e.,
to the ions [M ꢁ CnH2n+1O ꢁ CnH2n]+). Important conclu-
sions about the structures of these by-products were
obtained by considering their RI values. For instance, the
variation Pr ! Bu (i.e., adding the homologous increment
CH2) increases the RI values of alkyl diazoacetates (I),
chloroacetates (III), and nitroglycolates (IV) by approx. 100
i.u., in accordance with the RI concept.[8] However, the same
structural transformation (Pr ! Bu) increased the RI values
of compounds VI by approx. 200 i.u., indicating the presence
of two -CO2R structural fragments in the molecule, and
giving its structure as X(CO2R)2, where X is unknown.
To determine the molecular weight of these ’pseudo-dimeric’
by-products (VI) the PCI mass spectrum of compound VIb
(R = Et) was recorded using an Agilent 7000 GC-QQQ GC/MS
system using methane as the reagent gas. The [M + H]+
peak was found at m/z 204 (100 % relative intensity) and
satellite ion peaks were found at m/z 232 [M + C2H5]+ and
Acknowledgement
The authors are grateful to Dr. Gareth Thomas (Editing Science,
UK) for editing the English of the manuscript.
REFERENCES
[1] G. P. Barreto, A. I. Canizo, G. N. Eyler. Use of retention data as
the first step in the identification of cyclic organic peroxides in
temperature-programmed gas chromatography. Chromatogra-
phia 2006, 63, 261.
[2] B. P. Axe. Trace level detection and quantitation of ethyl diazoa-
cetate by reversed-phase high performance liquid chromato-
graphy and UV detection. J. Pharm. Biomed. Anal. 2006, 41, 804.
Rapid Commun. Mass Spectrom. 2013, 27, 461–466
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

T. A. Kornilova et al.
[3] The NIST 11 Mass Spectral Library (NIST11/2011/EPA/NIH).
Software/Data Version (NIST08); NIST Standard
Reference Database, Number 69, National Institute of
Standards and Technology, Gaithersburg, MD 20899,
August, 2011. Available: http://webbook.nist.gov (accessed
September 2012).
[4] G. Paulett, R. Ettinger. Mass spectra and appearance
potentials of diazirine and diazomethane. J. Chem. Phys.
1963, 39, 825.
[5] D.S. Wulfman, J. Roberts, D. K. Henderson, J. C. Romine,
R. S. McDaniel, L. Carver, D. W. Beistel. Metal salt catalysed
carbenoids. XVIII. The electron impact mass spectral
behavior of diazocarbonyl compounds. I. Can. J. Chem.
1984, 62, 554.
[6] A. T. Lebedev, T. Yu. Samguina, V. S. Petrosyan,
T. Blumenthal, A. M. Sipyaguin, V. G. Kartsev. Collision-
induced dissociation study of cyclization of a-diazo-o-
arylsulfonylaminoalkan-2-ones. Org. Mass Spectrom. 1992,
27, 730.
[8] I. G. Zenkevich, in Encyclopedia of Chromatography, (3rd edn.),
(Ed: J. Cazes). Taylor and Francis. New York, 2010,
pp. 1304–1310.
[9] S. E. Stein, V. I. Babushok, R. L. Brown, P. J. Linstrom.
Estimation of Kovats retention indices using group contri-
butions. J. Chem. Inf. Model. 2007, 47, 975.
[10] Weigand-Hilgetag. Experimental Methods in Organic Chemistry
(in Russian). Khimia, Moscow, 1968.
[11] Diazoacetophenone. Organic Syntheses 1973, 53, 35; 1988, 6,
368.
[12] M. C. Hamming, N. G. Foster. Interpretation of Mass Spectra
of Organic Compounds. Academic Press, New York, 1979.
[13] B. Akhlaghinia, S. Samiei. A novel and highly selective
conversion of alcohols, thiols, and silyl ethers to azides
using the 2,4,6-trichloro[1,3,5]triazine/n-Bu4NN3 system.
J. Braz. Chem. Soc. 2007, 18, 1311.
[14] O. M. Nefedov, A. I. Ioffe, L. G. Menchikov. Chemistry of
Carbenes (in Russian). Khimia, Moscow, 1990.
[15] U. Schollkopf, P. Tonne, H. Schafer, P. Markusch. Synthesis
von Nitro-diazoessigsaureestern, Nitro-cyano- und Nitro-
trifluoromethyl-diazomethane. Ann. Chem. 1968, 722, 45.
[7] A. T. Lebedev. Mass spectrometry of diazo compounds.
Mass Spectrom. Rev. 1991, 10, 91.
wileyonlinelibrary.com/journal/rcm
Copyright © 2012 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2013, 27, 461–466