6884 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Dyke et al.
(1:1) to pH ) 5. The product, N3CH2CO2H, was then extracted with
diethyl ether and dried over anhydrous MgSO4. Final purification to
remove diethyl ether and the last traces of water required prolonged
pumping under vacuum (10-6 mbar) at room temperature. Sample
but above the photon beam. The tantalum insert was heated to ca.
1000 K by radiofrequency induction heating,13,14 and the stainless steel
furnace containing the azidoacetic acid was heated by radiation from
the ceramic insert. Although this system provided a high enough
temperature to obtain full pyrolysis of the parent azide, it was found
that sufficiently controlled pyrolysis could not be achieved.
1
purity was estimated to be >99% on the basis of H NMR and mass
spectrometric data, a summary of which is included in the Results.
Routine precautions were taken to minimize the effect of possible
explosions at all stages in the preparation and handling of azide mate-
rials, but no untoward occurrences were experienced during this work.
Matrix Isolation IR Spectroscopy. Matrix isolation studies on the
thermal decomposition of azidoacetic acid were based around a
conventional closed-cycle cooling unit (APCI “Displex”, model
CSW202) as described elsewhere.12 For these particular studies, the
parent azidoacetic acid vapor was admitted to the system via a PTFE
needle valve at a pressure monitored by a Pirani gauge. The vapor
then passed through a 15 cm length of 5 mm i.d. silica tubing connected
directly to the housing of the Displex. This provided a direct line-of-
sight to a central CsI deposition window, which was maintained at ca.
12 K. Vapors passing through this silica tube could be superheated to
ca. 1000 K, by a concentric cylindrical furnace, and subsequently co-
condensed on the central cooled window12 with a large excess
(>×1000) of matrix gas.
A new pyrolysis system using a resistively heated furnace was there-
fore developed to allow controlled superheating over longer times. In
this inlet system, molybdenum wire was wrapped around an inner glass
tube and connected via tungsten feedthroughs in an outer glass tube to
a 15 V dc power supply with a maximum operating current of 2 A.
Above the heating wire, the glass tube widened and sample vials were
placed on a ledge in this section which was not heated directly. It was
found that the maximum operating temperature reached by this system
was 850 K in the heated section (as measured by a chromel-alumel
thermocouple). The upper region of the furnace containing vials of
azidoacetic acid was heated by conduction, and the temperature reached
was found to be high enough to generate vapor pressures of azidoacetic
acid which gave photoelectron spectra with good signal-to-noise ratios.
Furthermore the operating resolution of the photoelectron spectrometer
(typically 30-35 meV as measured for argon (3p)-1 fwhm ionized with
He IR radiation) was found to be unaffected by this heating system
even at the highest temperature available (ca. 850 K).
Photoelectron spectra were recorded in real time as the furnace
temperature was varied in the temperature range 300-850 K. The onset
of decomposition was marked by the appearance of characteristic PES
bands associated with molecular N2 and CO2. Spectral calibration was
achieved at low ionization energy (9.0-11.0 eV) by CH3I and at higher
energies by reference to the known ionizations of (added) Ar or traces
of H2O present in the system.15,16
In a typical experiment, a sample of azidoacetic acid was pre-cooled
to ca. 281-283 K prior to admission to provide better flow control
and was subsequently co-condensed with nitrogen as the inert matrix
gas. This host was chosen in preference to argon, in order to provide
a more uniform environment for any trapped species, as it was expected
that molecular nitrogen would be generated during thermal decomposi-
tion.
The initial deposition was usually carried out with the superheater
switched off, in order to record the matrix IR spectrum of the parent
azidoacetic acid and to establish a suitable flow rate. Several spectra
were then recorded after periods of deposition with the superheater set
at increasingly higher temperatures, in the anticipation that at some
stage new features arising from decomposition products would be
isolated in the matrix. From time to time, the superheater temperature
was reduced to room temperature to check that the flow of parent
azidoacetic acid remained unchanged. Typical deposition times
between spectra varied from 30 to 90 min.
PE spectra were also obtained for BrCH2CO2H, used in the
preparation of HCO2CH2N3. The lowest energy ionization of this
molecule consists of an intense, closely spaced doublet at 10.61 and
10.95 eV. This feature is assigned to spin-orbit coupling, and the
separation (0.34 eV) is typical of a Br-localized ionization.17 However,
no evidence for BrCH2CO2H was found in photoelectron spectra
recorded for purified samples of azidoacetic acid.
Results
Matrix IR spectra were recorded over the frequency range 4000-
300 cm-1 using either a Perkin-Elmer 983G or a Bio-Rad FTS60
spectrometer equipped with routine spectral subtraction procedures, and
any spectral changes taking place as a result of changes in the inlet
temperature or sample flow rate could readily be detected.
Characterization of N3CH2CO2H. Samples of purified
azidoacetic acid were characterized in the vapor phase by mass
spectrometry and UV photoelectron spectroscopy and in the
1
condensed phase by H NMR, Raman, and IR spectroscopy.
In addition to monitoring changes in IR band intensity during
deposition, it was also possible to investigate the effect of controlled
diffusion on the trapped species. During these studies, the temperature
of the matrix was cycled within the temperature range ca. 12-35 K.
Separate matrix isolation studies were also carried out to confirm
the positions of the IR bands of H2O, CO, CO2, HCN, and HCOOH in
N2 matrices and to establish, for the purpose of elimination, the matrix
IR spectra of BrCH2CO2H and the Et2O solvent used in the preparation
of the azidoacetic acid.
The 70 eV electron impact mass spectrum showed a prominent
parent ion peak at 101 amu (55%), together with two intense
fragments at 28 (97% N2+) and 45 (100%, CO2H+) amu. The
other signals observed were all significantly less intense but
included a very weak doublet at 138/140 amu (ca. 1%) arising
from unreacted BrCH2CO2H.
Only two signals of any significance were observed in the
1
300 MHz H NMR spectrum of azidoacetic acid in CDCl3
solution. These were observed at 4.0 and 10.3 ppm relative to
TMS, and their relative positions and integrated intensities (ca.
1.7:1) are consistent with their assignment as methylene and
hydroxyl protons, respectively.
The most intense absorption in the (liquid) IR spectrum was
found at 2117 cm-1, and this was assigned to a vibration of the
N3 group. Other prominent bands were present at 1700-1800
(br, C-O stretch), 1421, 1285, and 1250-1150 (br) cm-1. The
liquid Raman spectrum exhibited counterparts to all the above
bands (2116, 1730 (br), 1422, 1285, and 1225 cm-1) and in
addition showed intense features of 2921 (C-H stretch), 947,
880, 236, and 171 cm-1. The O-H stretching region showed
Photoelectron Spectroscopy. All photoelectron spectra recorded
in this work were obtained using He(I) (21.22 eV) radiation. The
spectrometer used was a single-detector instrument designed for high-
temperature work13,14 with a modified inlet system capable of super-
heating a vapor species immediately prior to photoionization. The vapor
pressure of the parent material, N3CH2CO2H, was insufficient at room
temperature to allow photoelectron spectra to be recorded with
acceptable signal-to-noise. Samples were therefore preheated to ca.
373-423 K in order to generate a greater partial pressure to allow
spectra with greater signal-to-noise ratio to be recorded.
In initial thermal decomposition experiments, a sample of liquid N3-
CH2CO2H was placed in a small stainless steel furnace inside the
spectrometer. A short length of cylindrical ceramic tube with a thin
tantalum inner liner was placed approximately 2 cm beneath the furnace
(15) Turner, D. W.; Baker, C.; Baker, A. D.; Brundle, C. R. Molecular
Photoelectron Spectroscopy; Wiley Interscience: New York, 1971.
(16) Kimura, K.; Katsumata, S.; Achiba, Y.; Yamaszaki, T.; Iwata, S.
Handbook of HeI Photoelectron Spectra; Japan Scientific Press: Tokyo,
1981.
(17) Eland, J. H. D. Photoelectron Spectroscopy; Butterworths: London,
1974.
(12) See, e.g.: Ogden, J. S.; Wyatt, R. S. J. Chem. Soc., Dalton Trans.
1987, 859.
(13) Morris, A.; Dyke, J. M.; Josland, G. D.; Hastings, M. P.; Francis,
P. D. High Temp. Sci. 1986, 22, 95.
(14) Bulgin, D.; Dyke, J. M.; Goodfellow, F.; Jonathan, N. B. H.; Lee,
E.; Morris, A. J. Electron Spectrosc. Relat. Phenom. 1977, 12, 67.