N-methyl-4-hydrazino-7-nitrobenzofurazan as a new reagent for air monitoring of aldehydes and ketones

Article abstract of DOI:10.1021/ac980946f

The synthesis of N-methyl-4-hydrazino-7-nitrobenzofurazan (MNBDH) and its application as a new reagent for the determination of aldehydes and ketones are described. MNBDH reacts with carbonyl compounds in acidic media to the corresponding MNBD-hydrazones.

Full text of DOI:10.1021/ac980946f

Anal. Chem. 1999, 71, 1893-1898
N-Me t h yl-4 -h yd ra zin o -7 -n it ro b e n zo fu ra za n a s a Ne w
Re a g e n t fo r Air Mo n it o rin g o f Ald e h yd e s a n d
Ke t o n e s
Andre a B u1 ldt a nd Uw e Ka rs t*
Abteilung Analytische Chemie, Anorganisch-Chemisches Institut, Westf a¨ lische Wilhelms-Universit a¨ t M u¨ nster,
Wilhelm-Klemm-Strasse 8, 48149 M u¨ nster, Germany
The synthesis of N-methyl-4 -hydrazino-7 -nitrobenzofura-
zan (MNBDH) and its application as a new reagent for
the determination of aldehydes and ketones are de-
scribed. MNBDH reacts with carbonyl compounds in
acidic media to the corresponding MNBD-hydrazones. In
contrast to the established reagent 2 ,4 -dinitrophenylhy-
drazine (DNP H), MNBDH is oxidized by both ozone and
nitrogen dioxide quantitatively to only one product, N-meth-
yl-4-amino-7-nitrobenzofurazan (MNBDA). This can easily
be separated from the hydrazones of lower aldehydes by
means of HP LC. Due to larger molar absorptivities and
absorption maxima at wavelengths over 4 70 nm, selectiv-
ity is higher and limits of detection are lower for the new
reagent compared to DNP H. MNBDH reacts slightly faster
than DNP H with carbonyl compounds and significantly
faster than other N-alkylated hydrazine reagents.
mination of aldehydes and ketones in air samples due to oxidative
decomposition of the reagent and the hydrazones, respec-
tively. In case of ozone, several products are formed, but only
some of those have been identified so far.10 The reaction of DNPH
with nitrogen dioxide leads to 2,4-dinitrophenyl azide (DNPA).
The above-mentioned problems are caused by coelution of
these reaction products with the hydrazones in chroma-
tography.
An approach to overcome this problem was described in ref
1
1. N-Methyl-2,4-dinitrophenylhydrazine (MDNPH) was used as
a new derivatization reagent with reduced interferences, because
the reaction of MDNPH with both nitrogen dioxide and ozone
led to only one defined product, N-methyl-2,4-dinitroaniline
(
MDNA). This compound could be separated from the hydra-
zones by liquid chromatography. However, a major disadvantage
of this new reagent is its reduced reactivity compared to that of
DNPH.
Air monitoring of aldehydes and ketones frequently employs
derivatization with an aromatic hydrazine reagent. Stable hydra-
zones are formed, which are separated by HPLC and detected by
UV/ visible spectroscopy. Among these reagents, 2,4-dinitrophe-
nylhydrazine (DNPH) is the most widely used. Several national
and international standardization bodies suggest this reagent for
the analysis of aldehydes and ketones in gaseous samples.
For this reason, a new hydrazine reagent should be developed
which combines the above-mentioned advantages concerning
the more defined reactivity toward nitrogen dioxide and ozone
with a fast derivatization reaction known from the DNPH
method.
EXPERIMENTAL SECTION
Sampling of carbonyl compounds in air can be performed using
solutions of DNPH in impingers1,2 or solid sorbents coated with
Chemicals. All chemicals were purchased from Aldrich
Chemie (Steinheim, Germany) in the highest quality available.
Acids were Merck (Darmstadt, Germany) analytical grade. Tri-
ethylamine was from Fluka (Neu-Ulm, Germany). Acetonitrile for
HPLC was Merck gradient grade.
3
DNPH, including test tubes for pumped sampling and passive
sampling devices.4,5 However, problems with respect to the
determination of aldehydes and ketones in complex matrixes were
described recently. Oxidants such as nitrogen dioxide or ozone
frequently occur in conjunction with carbonyl compounds, e.g.,
in ambient air or in automobile exhausts.2,6 The presence of
these substances7 causes problems regarding an exact deter-
1
Instrumentation for P roduct Identification. H NMR mea-
surements were performed with an AC 200 spectrometer (200.13
MHz) from Bruker (Bremen, Germany). The solvent for all NMR
-9
3
measurements was CDCl . All peaks are given as δ in ppm. FT-
(
(
(
1) Grosjean, D.; Fung, K. Anal. Chem. 1 9 8 2 , 54, 1221-1224.
2) Lipari, F.; Swarin, S. J. J. Chromatogr. 1 9 8 2 , 247, 297-306.
3) Beasley, R. H.; Hoffmann, C. E.; Rueppel, M. L.; Worley, J. W. Anal. Chem.
IR spectral information was obtained for the products in KBr
pellets by using an IFS48 instrument from Bruker. Mass spectra
were recorded on a MAT 212 from Varian. Elementary analyses
were performed with a Heraeus (Hanau, Germany) CHN-O-Rapid
1
9 8 0 , 52, 1110-1114.
4) Levin, J. O.; Andersson, K.; Lindahl, R.; Nilson, C.-A. Anal. Chem. 1 9 8 5 ,
7, 1032-1035.
(
5
(
(
(
5) Levin, J. O.; Lindahl, R. Analyst 1 9 9 4 , 119, 79-83.
6) Swarin, S. J.; Lipari, F. J. Liq. Chromatogr. 1 9 8 3 , 6, 425-444.
7) Karst, U.; Binding, N.; Cammann, K.; Witting, U. Fresenius’ J. Anal. Chem.
(9) Smith, D. F.; Kleindienst, T. U.; Hudgens, E. E. J. Chromatogr. 1 9 8 9 , 483,
431-436.
1
9 9 3 , 345, 48-52.
(10) Achatz, S. Diploma Thesis, Technical University of Munich, 1996.
(
8) Arnts, R. R.; Tejada, S. B. Environ. Sci. Technol. 1 9 8 9 , 23, 1428-1430.
(11) B u¨ ldt, A.; Karst, U. Anal. Chem. 1 9 9 7 , 69, 3617-3622.
1
0.1021/ac980946f CCC: $18.00 © 1999 American Chemical Society
Analytical Chemistry, Vol. 71, No. 9, May 1, 1999 1893
Published on Web 04/03/1999

instrument. The analytical data for the reagent and its derivatives
are provided as Supporting Information.
fully characterized in the same way as MNBDH and its hydra-
zones. These data are available as Supporting Information.
-
3
Syntheses. (1) N-Methyl-4-hydrazino-7-nitrobenzofurazan
P reparation of the MNBDH Sampling Tubes. A 8 × 10
(
MNBDH). A 2.24 mL sample of methylhydrazine (0.04 mol) in
00 mL of methanol was added dropwise to a solution of 1 g (0.005
mol/ L MNBDH solution was prepared by adding 100 mg of
MNBDH to 2 mL of concentrated sulfuric acid in 60 mL of
acetonitrile. The used sampling tubes (cartridge lengths 7.4 cm),
filled with 350 mg of silica gel (chromatographic quality; particle
size 150-250 µm, 60/ 100 mesh), were purchased as a special
order (part no. 8-54122) from Supelco (Deisenhofen, Germany).
These tubes were conditioned first with 3 mL of acetonitrile.
Afterward, the silica gel was coated with MNBDH by allowing 3
mL of the reagent solution to seep through the tubes. Subse-
quently, the material was dried in a nitrogen stream.
1
mol) of 4-chloro-7-nitrobenzofurazan in 80 mL of chloroform. After
being heated at reflux for 20 min, the solution was cooled, and
MNBDH precipitated as a red crystalline material. The precipitate
was filtered off and washed with methanol. The yield was 48%.
The purity of the product was examined by means of HPLC. The
1
product was fully characterized by means of H NMR, IR, UV,
MS, and elemental analysis. These data are provided as Supporting
Information.
(
2 ) Aldehyde and Ketone MNBD-Hydrazones. The deriva-
Air-Sampling P rocedure and Analysis. Air sampling was
performed using a personal air sampler pump (Buck I. H. Pump)
from A. P. Buck (Orlando, FL) with the corresponding calibrator
also from A. P. Buck. For all measurements, a collecting tube
and a controlling tube were connected in series to identify
incomplete recovery on the collecting tube. The sampling rate
was 1 L/ min. Sampling of the car exhaust was performed by
locating the collecting tube directly at the end of the exhaust pipe
of the car for 1 min. During the sampling of the air of the
disinfected room, the tubes were located in the middle of the room
for 2 min. The simulated air sampling with a defined amount of
tives were prepared according to a procedure published by
Behforouz et al.12 for the synthesis of DNPH derivatives. A 100
mg sample of N-methyl-4-hydrazino-7-nitrobenzofurazan (4.8 ×
-
4
1
0
mol) was dissolved in 0.7 mL of water, 0.5 mL of sulfuric
acid, and 2.5 mL of 95% ethanol. A 50% molar excess of the
-
4
aldehyde or ketone (7.2 × 10 mol) was added. The hydrazones
precipitated as reddish crystalline material. The precipitate was
filtered off and washed first with a 5 mass % aqueous solution of
sodium bicarbonate until no further development of carbon dioxide
was observed and then with distilled water. If necessary, the
products were recrystallized from ethanol. Yields ranged from 70%
to >95% depending on the degree of purification required for the
individual hydrazones. The reaction products were fully character-
-3
the analyte was performed by pipetting 50 µL of a 5.8 × 10 mol/ L
solution of formaldehyde in acetonitrile onto quartz wool that was
placed in front of the collecting layer. Afterward, a constant air
stream was pumped through the tube for 4 min. After sampling,
the tubes were eluted with 10 mL of a 1% solution of sulfuric acid
in acetonitrile. A 10 µL portion of this solution was injected into
the HPLC system.
1
ized by means of H NMR, IR, UV, MS, and elemental analysis.
These data are available as Supporting Information.
(
3) N-Methyl-4-amino-7-nitrobenzofurazan (MNBDA). (a)
Synthesis by Reaction with Nitrogen Dioxide or Ozone. A 100
mg sample of N-methyl-4-hydrazino-7-nitrobenzofurazan (4.7 ×
-
4
P hotometer. An HP 8453 diode array spectrophotometer
1
0
mol) was dissolved in 2.5 mL of distilled water, 0.8 mL of
(
Hewlett-Packard, Waldbronn, Germany) with HP Chem Station
sulfuric acid, and 0.3 mL of acetonitrile. Nitrogen dioxide gas was
produced by adding 10 mL of concentrated nitric acid to 2 g of
copper wire and transported into the reagent solution by nitrogen
carrier gas. Alternatively, ozone from an ozone generator was
transported into the reagent solution by oxygen carrier gas. After
the reaction, nitrogen was blown into the solutions to remove
unreacted oxidant. The precipitate was filtered off and washed
with sodium bicarbonate solution (5%) as well as with distilled
water.
845x-biochemical UV/ VIS-system software was used.
UV/ Visible Absorption Measurements. All UV/ visible mea-
surements were performed in the concentration range from 8.1
-
5
-5
× 10 to 5.3 × 10 mol/ L for MNBDH, MNBDA, and the
hydrazones in acetonitrile. The spectra were recorded in the range
from 290 to 600 nm.
Spectrofluorophotometer. An RF-5301 PC spectrofluoropho-
tometer (Shimadzu, Duisburg, Germany) with software version
1.10 was used.
(
b) Synthesis According to a P rocedure P ublished by
Clusius and Schwarzenbach.13 A 200 mg sample of N-methyl-
HP LC Instumentation and Analysis. A high-performance
liquid chromatograph consisting of the following components was
used: two LC-10AS pumps (Shimadzu, Duisburg, Germany), SPD-
M10Avp diode array detector (Shimadzu), SIL-10A autosampler
-
4
4
-hydrazino-7-nitrobenzofurazan (9.4 × 10 mol) was dissolved
in 30 mL of ethanol and 5 mL of concentrated hydrochloric acid.
While the mixture was cooled with ice, a solution of 70 mg of
-3
sodium nitrite (1 × 10 mol) in 5 mL of distilled water was added.
After the reaction solution was allowed to stand for 0.5 h at room
temperature, 50 mL of distilled water was added. Afterward, the
solution was left at room temperature again for 1 h. The product
precipitated as a yellow material. This was filtered off and washed
with ice-cold water. The yield was 67%. The reaction product was
(
Shimadzu), Class LC-10 version 1.6 software (Shimadzu), and
CBM-10A controller unit (Shimadzu). The injection volume was
0 µL. The column material was Merck LiChroSpher RP-18
Merck, Darmstadt, Germany) in ChromCart cartridges (Mach-
1
(
erey-Nagel, D u¨ ren, Germany): particle size 5 µm; pore size 100
Å; column dimensions 250 mm × 3 mm; guard column 8 mm ×
3
mm.
(
(
12) Behforouz, M.; Bolan. J. L.; Flynt, M. S. J. Org. Chem. 1 9 8 5 , 50, 1186-
189.
13) Clusius, K.; Schwarzenbach, K. Helv. Chim. Acta 1 9 5 8 , 41, 1413-1416.
For separation, binary gradients consisting of acetonitrile and
a mixture of water/ triethylamine/ acetic acid (preparation: 500
1
1894 Analytical Chemistry, Vol. 71, No. 9, May 1, 1999

Ta b le 1 . UV/Vis ib le S p e c t ro s c o p ic a n d Ch ro m a t o g ra p h ic Da t a fo r a S e rie s o f MNBD-Hyd ra zo n e s , MNBDH, a n d
MNBDA
MNBD-hydrazone of
10-3ꢀ(474 nm)
10-3ꢀ(514 nm)
ꢀ(474 nm)/ ꢀ(514 nm)
A(474 nm)/ A(514 nm)
λmax (nm)
10-3ꢀ(λmax)
formaldehyde
acetaldehyde
propanal
acetone
acrolein
24.7
25.7
25.0
18.4
23.0
24.0
21.2
19.9
27.8
24.9
25.3
25.2
23.4
26.3
38.2
24.5
20.2
20.2
20.0
6.48
20.0
20.3
21.3
25.6
19.9
25.2
31.6
23.6
21.3
21.7
21.6
20.5
22.5
32.2
36.1
35.2
11.1
1.0
3.9
1.3
1.2
0.9
0.9
1.2
0.8
0.6
1.2
1.2
1.2
1.2
1.2
1.2
1.2
0.7
0.6
1.8
20.0
2.5
1.0
1.0
0.8
0.7
1.0
0.7
0.6
1.1
1.1
1.1
1.2
1.2
1.2
1.2
0.5
0.5
1.1
10.6
474
488
492
500
499
492
501
511
492
492
493
494
493
493
492
508
514
486
459
24.7
29.0
28.5
24.6
28.8
27.7
28.6
31.8
31.7
28.9
29.5
27.6
27.3
30.7
44.2
33.9
34.4
21.6
23.3
butanal
2-butanone
crotonaldehyde
pentanal
hexanal
heptanal
octanal
nonanal
decanal
glutaraldehyde
benzaldehyde
p-tolualdehyde
MNBDH
MNBDA
mL of water with 2415 µL of triethylamine and 975 µL of acetic
acid, pH ≈ 7.5) were chosen with the following profile:
The carbonyl compounds react with MNBDH in the presence of
sulfuric acid, forming the corresponding hydrazones:
Neither the reagent nor any of the respective hydrazones have
been described in the literature before and are now applied to
the determination of aldehydes and ketones in air samples. The
hydrazones are separated by means of high-performance liquid
chromatography (HPLC) and detected by UV/ visible spectroscopy
at wavelengths which depend on the absorption maxima of the
relevant hydrazones. In Table 1, the UV/ visible spectroscopic data
measured photometrically are shown for MNBDH and a series
of hydrazones. Compared to the spectra of the DNPH derivatives,
a shift of the absorption maxima of 100-150 nm is observed. This
large shift may lead to higher selectivity of the new method
especially in the presence of colored matrix constituents. Fur-
thermore, the molar absorptivities of the derivatives are higher
compared to those of the derivatives of DNPH, thus resulting in
lower limits of detection. Table 2 shows the comparison of the
detection limits of several MNBD-hydrazones with those of DNP-
hydrazones that have been obtained by means of HPLC with UV/
visible detection under identical elution conditions but using the
respective absorption maxima for quantification. A fluorimetric
detection after HPLC separation is possible for the MNBD-
RESULTS
The derivatization of carbonyl compounds with 4-hydrazino-
7
-nitrobenzofurazan14,15 is known from the literature. Although a
UV/ visible detection at significantly longer wavelengths compared
to those of the DNPH method is possible, the reagent has yet
found only limited use. This is, on one hand, due to the formation
of a hydrazine adduct of the reagent during synthesis. On the
other hand, only the analysis of longer-chained aliphatic aldehydes
3
starting with C has been described in the literature and not the
analysis of the most important aldehydes, especially formalde-
hyde.
To use the excellent chromophoric properties of these com-
pounds, the N-methylated hydrazine appeared to be an attractive
alternative. N-Methyl-4-hydrazino-7-nitrobenzofurazan (MNBDH)
was synthesized in a nucleophilic substitution reaction of N-
methylhydrazine with 4-chloro-7-nitrobenzofurazan:
(
(
14) G u¨ bitz, G.; Wintersteiger, R.; Frei, R. W. J. Liq. Chromatogr. 1 9 8 4 , 7, 839-
54.
15) Uzu, S.; Kanda, S.; Imai, K.; Nakashima, K.; Akiyama, S. Analyst 1 9 9 0 , 115,
477-1482.
8
1
Analytical Chemistry, Vol. 71, No. 9, May 1, 1999 1895

Ta b le 2 . Co m p a ris o n o f t h e De t e c t io n Lim it s o f S e ve ra l
MNBD-Hyd ra zo n e s w it h Th o s e o f DNP -Hyd ra zo n e s
detection limit
detection limit
-
8
-8
(
10 mol/ L)
(10 mol/ L)
using using
DNPH MNBDH
using
using
DNPH MNBDH
formaldehyde
propanal
crotonaldehyde
6
5
4
3
3
2
acetone
p-tolualdehyde
4
4
2
2
hydrazones as well. They show an excitation maximum of about
70 nm, an emission maximum of about 560 nm, and a resulting
4
Stokes shift of 90 nm. Unfortunately, the fluorimetric detection
does not lead to improved limits of detection compared to the
UV/ visible detection due to the weak fluorescence of the hydra-
zones. However, fluorescence detection may provide higher
selectivity when carbonyls are determined in complex matrixes.
The reaction of MNBDH with ozone and nitrogen dioxide was
investigated. MNBDH reacts with both oxidants, resulting in
almost quantitative formation of one single product, N-methyl-4-
amino-7-nitrobenzofurazan (MNBDA):
Fig u re 1 . Chromatogram of a series of aldehyde and ketone MNBD-
hydrazones, MNBDH, and MNBDA at the detection wavelengths λ
)
474 nm (solid) and 514 nm (dotted) using gradient A. The
concentration range of the compounds dissolved in acetonitrile was
5
(
1.1-4.5) × 10- mol/L.
HPLC coincide well. The variations may be traced back to the
reasons discussed in ref 16 and to the use of different mobile
phases.
Another interesting aspect regarding the chromatogram in
Figure 1 is the elution order of the hydrazones. It corresponds to
the elution order of other N-alkylated hydrazones as described
in ref 11 and exhibits significant differences compared to the
elution order observed for the DNP-hydrazones. The hydrazones
of the aliphatic aldehyde hydrazones elute in order of increasing
alkyl chain lengths, while the ketone hydrazones elute significantly
in advance of aldehyde hydrazones with even shorter alkyl chain
lengths. Furthermore, it should be noted that MNBDA can easily
be separated from the MNBD-hydrazones, thus avoiding interfer-
ences from a coelution as in case of DNPH.
As a very positive aspect, MNBD-hydrazones are not decomposed
by nitrogen dioxide or ozone as long as an excess of MNBDH is
present. The UV/ visible spectroscopic properties of MNBDA are
similar to those of MNBDH and the MNBD-hydrazones and are
listed in Table 1.
First, HPLC chromatograms of the new reagent on reversed-
phase columns with a binary water/ acetonitrile gradient yielded
not one but several peaks in the chromatogram. After addition of
a carbonyl compound, all of these peaks disappeared and only
one new peak of the hydrazone was formed. This indicated that
equilibria between the reagent and solvent molecules or between
at least two reagent molecules lead to the formation of stable
intermediates which may be separated in HPLC. To suppress
these equilibria, acetic acid and triethylamine (for concentrations,
see Experimental Section) were added to the aqueous eluent. As
can be seen in Figure 2, significant differences are observed: If
only pure water and acetonitrile are used for the gradient, several
peaks appear in the chromatogram. If the additives are used, only
one peak appears in the chromatogram. It should be noted that
the areas of the reagent peaks without using the additives varied
strongly between different chromatograms, even when these were
recorded within only a short time span. All separations should
therefore be carried out using the additives to avoid misinterpreta-
tions of the additional peaks.
The differences concerning the absorption maxima of the
substances can be used to identify groups of aldehyde and ketone
hydrazones by dual-wavelength detection after the HPLC separa-
tion. As the absorption spectra of formaldehyde MNBD-hydrazone
and p-tolualdehyde MNBD-hydrazone are rather different, the
respective absorption maxima at λ ) 474 and 514 nm are selected
for this purpose.
Figure 1 presents a chromatogram of several MNBD-hydra-
zones, MNBDH, and MNBDA recorded at these wavelengths.
While MNBDH and the aliphatic aldehyde MNBD-hydrazones
exhibit a ratio (ꢀ(λ ) 474 nm)/ ꢀ(λ ) 514 nm)) of approximately
1
, it is obvious that the ratio for formaldehyde MNBD-hydrazone
as well as for MNBDA is much larger than 1. The ketone and
unsaturated aldehyde hydrazones show a ratio slightly smaller
than 1. Furthermore, it is obvious that the aromatic aldehyde
hydrazones are characterized by a ratio much smaller than 1.
These differences show very clearly that the dual-wavelength
detection can be a very helpful tool for the identification of different
groups of carbonyl compounds in HPLC. The quotients of the
peak areas at these two wavelengths are listed in Table 1 besides
the results given by photometric investigations. In most cases,
the values obtained by UV/ visible spectrophotometry and by
As the major drawback of the only known N-alkylated hydra-
zine reagent MDNPH was its slow reaction with the carbonyl
compounds, special focus should be directed to the reactivity of
(16) P o¨ tter, W.; Karst, U. Anal. Chem. 1 9 9 6 , 68, 3354-3358.
1896 Analytical Chemistry, Vol. 71, No. 9, May 1, 1999

Fig u re 4 . Absorbance/time curves for the formation of formaldehyde
-
7
DNP-hydrazone in acetic solution (6.5 × 10 mol of DNPH in a
mixture of 1.4 mL of H2SO4, 1.9 mL of H2O, and 6.7 mL of ethanol)
at the wavelength of 355 nm (star) and the consumption of DNPH at
295 nm (circle).
Fig u re 2 . Chromatogram of MNBDH analyzed with different mobile
phases using gradient B: acetonitrile and water (dotted) and aceto-
nitrile and water/triethylamine/acetic acid (solid). The concentration
was 9.7 × 10- mol/L in acetonitrile.
5
Fig u re 5 . Chromatogram of a car exhaust sample at the detection
wavelength λ ) 474 nm using gradient A: sampling layer (dotted);
controlling layer (solid).
reaction of MDNPH with formaldehyde, however, requires several
hours. Therefore, MNBDH can be considered to be an excellent
alternative to DNPH with regard to reactivity.
Fig u re 3 . Absorbance/time curves for the formation of formaldehyde
MNBD-hydrazone in acetic solution (5.8 × 10-7 mol of MNBDH in a
mixture of 1.4 mL of H2SO4, 1.9 mL of H2O, and 6.7 mL of ethanol)
at the wavelength of 474 nm (star) and the consumption of MNBDH
at 400 nm (circle).
In the following, MNBDH was used to prepare test tubes for
air sampling of the carbonyls as stated above. The recovery rate
of formaldehyde was determined by pipetting a solution with a
defined amount of formaldehyde in acetonitrile (50 µL of a 5.8 ×
-
3
the new reagent MNBDH. The investigations were carried out
with UV/ visible spectroscopy. Special experimental conditions had
to be chosen according to the reasons mentioned in ref 11. For
the measurements, an excess of the carbonyl compound was
added directly to an acidified reagent solution in a cuvette. Figure
10 mol/ L solution) onto quartz wool which was placed in front
of the collecting layer. Afterward, a constant air stream of 1 L/ min
was pumped through the tubes. Two test tubes were connected
in series to observe possible incomplete recoveries on the
collecting tube. Both collecting and controlling tubes were
analyzed by means of HPLC. This way, the recovery rate of
formaldehyde was determined 10-fold with a formaldehyde amount
of 8.7 µg. In all cases, no breakthrough of the analyte in the
controlling layer could be observed. The average recovery was
97.5% with a standard deviation of 1.5%. These investigations prove
the suitability of MNBDH-coated test tubes for the determination
of carbonyl compounds in gaseous samples.
3
shows the absorbance/ time curves of the reaction of MNBDH
with formaldehyde. The consumption of the protonated reagent
can be observed at λ ) 400 nm, while the formation of formal-
dehyde MNBD-hydrazone is recognized at its absorption maxi-
mum (λ ) 474 nm). The reaction is completed after 150 s. To
compare these results with those for DNPH, the reaction of DNPH
with formaldehyde, carried out under the same conditions, is
shown in Figure 4. The consumption of protonated DNPH is
observed at λ ) 295 nm and the formation of the hydrazone at λ
Next, real samples were analyzed using the MNBDH-coated
test tubes. A car exhaust sample was selected as an example for
a matrix with a very high content of nitrogen dioxide. Figure 5
)
355 nm. This reaction is completed after 250 s. The quantitative
Analytical Chemistry, Vol. 71, No. 9, May 1, 1999 1897

mediately after disinfection. This value agrees very well with the
amount of formaldehyde found by means of the DNPH method
(
1.4 ppm).
The investigations of the car exhaust sample and the disin-
fected room sample prove the suitability of the new reagent
MNBDH for the determination of carbonyl compounds and
nitrogen dioxide in real air samples.
CONCLUSIONS
MNBDH has been synthesized and applied as a new reagent
for the determination of aldehydes and ketones with several
advantages compared to the well-known DNPH method. Reagent
and derivatives are characterized by high absorption maxima and
molar absorptivities. Due to this fact, the selectivity is higher and
the limits of detection are lower compared to those of the
established reagent. The reaction of MNBDH with both nitrogen
dioxide and ozone leads to only one product, MNBDA, which
offers the possibility of determining carbonyl compounds simul-
taneously with nitrogen dioxide in real samples. MNBDH, MN-
BDA, and the hydrazones can be separated easily by means of
reversed-phase HPLC using UV/ visible detection. The reaction
of MNBDH with carbonyl compounds has been investigated by
means of UV/ visible spectroscopy and found to be even slightly
faster compared to the reaction of DNPH with the respective
carbonyl compounds.
Fig u re 6 . Chromatogram of an air sample of a disinfected room at
the detection wavelength of λ ) 474 nm using gradient B: sampling
layer (dotted); controlling layer (solid).
shows the corresponding chromatogram of the car exhaust with
a volume of 1 L from a diesel-fueled car. Data for both collecting
and controlling layers are presented in this figure. Again, there
was no breakthrough of the analytes in the controlling layer. As
is obvious from the large MNBDA peak, the exhaust contains
significant amounts of nitrogen dioxide. Other constituents of the
exhaust are formaldehyde and acetaldehyde. The measurement
was carried out twice. The concentrations of the pollutants found
in these two measurements were 11.2 and 11.3 ppm of nitrogen
dioxide, 2.0 and 2.1 ppm of formaldehyde, and 0.32 and 0.33 ppm
of acetaldehyde, respectively. It should be noted that the data
represent only the concentrations of the pollutants in the exhaust
sample. As the total exhaust volume is not known, this cannot be
correlated to the total amount of emitted aldehydes from the
automobile. The calculations concerning the amount of nitrogen
dioxide are only allowed on the supposition that the MNBDA peak
is exclusively caused by reaction of MNBDH and nitrogen dioxide.
Additionally, an air sample was taken from a room which had
been subjected to disinfection with a formaldehyde-containing
disinfectant. An air volume of 2 L was pumped through the tubes
as described above. In Figure 6, the chromatograms of both the
collecting layer and the controlling layer are presented. Again,
no breakthrough of the aldehydes in the controlling layer was
observed. The sample contained 1.4 ppm of formaldehyde im-
ACKNOWLEDGMENT
Financial support from the Fonds der Chemischen Industrie
(
Frankfurt, Germany) is gratefully acknowledged.
SUPPORTING INFORMATION AVAILABLE
1
Text presenting H NMR, MS, IR, and elemental analysis data
for N-methyl-4-hydrazino-7-nitrobenzofurazan, formaldehyde MNBD-
hydrazone, acetaldehyde MNBD-hydrazone, propanal MNBD-
hydrazone, butanal MNBD-hydrazone, pentanal MNBD-hydra-
zone, hexanal MNBD-hydrazone, heptanal MNBD-hydrazone,
octanal MNBD-hydrazone, nonanal MNBD-hydrazone, decanal
MNBD-hydrazone, acetone MNBD-hydrazone, 2-butanone MNBD-
hydrazone, acrolein MNBD-hydrazone, crotonaldehyde MNBD-
hydrazone, glutaraldehyde MNBD-hydrazone, benzaldehyde MN-
BD-hydrazone, p-tolualdehyde MNBD-hydrazone, and N-methyl-
4
-amino-7-nitrobenzofurazan. This material is available free of
charge via the Internet at http:/ / pubs.acs.org.
Received for review August 21, 1998. Accepted February
6, 1999.
AC980946F
1898 Analytical Chemistry, Vol. 71, No. 9, May 1, 1999