Simultaneous determination of C1-C4 carboxylic acids and aldehydes using 2,4-dinitrophenylhydrazine-impregnated silica gel and high-performance liquid chromatography

Article abstract of DOI:10.1021/ac0493471

A new method for the simultaneous determination of aliphatic carboxylic acids and aldehydes in air is described. In this work, carboxylic acids were allowed to react with 2,4-dinitrophenylhydrazine (DNPH) to form the corresponding carboxylic 2,4-dinitroph

Full text of DOI:10.1021/ac0493471

Anal. Chem. 2004, 76, 5849-5854
Simultaneous Determination of C1-C4 Carboxylic
Acids and Aldehydes Using
2,4-Dinitrophenylhydrazine-Impregnated Silica Gel
and High-Performance Liquid Chromatography
Shigehisa Uchiyama,*,† Erika Matsushima,† Shohei Aoyagi,‡ and Masanori Ando§
Division of Environmental Chemistry, National Institute of Health Sciences, 1-18-1 Kamiyoga, Setagaya-ku,
Tokyo 158-8501, Japan, and Faculty of Engineering, Chiba University, 1-33, Yayoicho, Inage-ku, Chiba City,
Chiba, 263-8522, Japan
also take place on a filter impregnated with KOH 4 or Na2CO3;5 a
solid sorbent, such as a silica gel;6 or a KOH-coated C18 cartridge.7
On the basis of the large diffusion coefficient differences between
gases and particles, KOH- and Na2CO3-coated denuders can be
used to collect gaseous carboxylic acids.8,9 The eluted solution
analysis can be performed by various chromatographic methods,
including ion chromatography,10 ion exclusion chromatography,11
high performance liquid chromatography (HPLC),12 capillary
electrophoresis,13 and gas chromatography (GC)4. Recently,
SPME-GC/ MS14 has been applied to the measurement of acetic
and formic acids in a museum environment.
A new method for the simultaneous determination of
aliphatic carboxylic acids and aldehydes in air is de-
scribed. In this work, carboxylic acids were allowed to
react with 2,4-dinitrophenylhydrazine (DNP H) to form the
corresponding carboxylic 2 ,4 -dinitrophenylhydrazides.
These derivatives have excellent thermal stability, with
melting points higher than those of the corresponding
hydrazones by 3 2 -5 0 °C. C1 -C4 carboxylic acid 2 ,4 -
dinitrophenylhydrazides exhibited maximum absorption
wavelengths of 3 3 1 -3 3 4 nm and molar absorption coef-
ficients of 1 .4 × 1 0 4 L/ mol/ cm. They were completely
separated by high-performance liquid chromatography
(HPLC) with an RP-Amide C16 column. Cartridges packed
with DNP H-coated silica particles (DNP H cartridge) were
used for sampling formic acid and aldehydes. Formic acid
was physically adsorbed on the silica particles as the first
step of the sampling mechanism. Gradual reaction with
DNP H followed. Formic acid reacted very slowly with
DNP H at room temperature (2 0 °C), but reacted com-
pletely at 8 0 °C over 4 h. In field measurements, the
sample air was drawn through a DNP H cartridge. After
sampling, the cartridges were heated at 8 0 °C for 5 h and
extracted with acetonitrile for HP LC analysis. Under these
optimized conditions, the LOD is 0 .4 ug/ m3 for an air
sample collected for 2 4 h at 1 0 0 mL/ min (1 4 4 L).
The most commonly used method to determine aldehydes and
ketones exploits the reaction of an amino (NH2) group with a
carbonyl (CdO) moiety to form a stable Schiff base (CdN).
Derivatization is very useful in analysis by gas chromatography
or high-performance liquid chromatography. Various derivatization
reagents have been used for the determination of aldehydes and
ketones in air, including O-(2,3,4,5,6-pentafluorobenzyl)hydroxy-
lamine,15 O-benzylhydroxylamine,16 2-diphenylacetyl-1,3-indandi-
one-1-hydrazone,17 5-dimethylaminonaphthalene-1-sulfohydrazide
(dansylhydrazine),18 N-methyl-4-hydrazino-7-nitrobenzofurazan,19
(2) Kumar, N.; Kulshrestha, U. C.; Khare, P.; Saxena, A.; Kumari, K. M.;
Srivastava, S. S. Atmos. Environ. 1 9 9 6 , 30, 3545-3550.
(3) Gronberg, L.; Shen, Y.; Jonsson, J. A. J. Chromatogr., A 1 9 9 3 , 655, 207-
215.
(4) Kawamura, K.; Ng, L. L.; Kaplan, I. R. Environ. Sci. Technol. 1 9 8 5 , 19,
1082-1086.
(5) Allen, A. G.; Miguel, A. H. Atmos. Environ. 1 9 9 5 , 29, 3519-3526.
(6) Hekmat, M.; Smith, R. G. Am. Ind. Hyg. Assoc. J. 1 9 9 1 , 52, 332-335.
(7) Grosjean, D.; Van-Neste, A.; Parmar, S. S. J. Liq. Chromatogr. 1 9 8 9 , 12,
3007-3017.
(8) Lawrence, J. E.; Koutrakis, P. Environ. Sci. Technol. 1 9 9 4 , 28, 957-964.
(9) Souza, S. R.; Vasconcellos, P. C.; Carvalho, L. R. F. Atmos. Environ. 1 9 9 9 ,
33, 2563-2574.
Carbonyl compounds, including carboxylic acids, aldehydes,
and ketones, are ubiquitous pollutants throughout the tropo-
sphere, and numerous analytical methods have been reported in
many papers. The first reported analysis of aliphatic carboxylic
acids was carried out using a bubbler in NaOH solution followed
by silica gel column chromatography titration.1 Many additional
methods followed. The most common collection procedures for
aliphatic carboxylic acid in air involve trapping in a liquid, such
as water2 or aqueous solutions of NaOH3. Sample collection may
(10) Johnson, B. J.; Huang, S. C.; Wong, A.; Yao, L. Microchem. J. 1 9 9 4 , 49,
79-84.
(11) Tanaka, K.; Fritz, J. S. J. Chromatogr. 1 9 8 6 , 361, 151-160.
(12) Vainiotalo, S.; Pfaffli, P.; Zitting, A. J. Chromatogr. 1 9 8 3 , 258, 207-221.
(13) Surowiec, K.; Dasgupta, P. K. J. Microcolumn Sep. 1 9 9 8 , 10, 265-271.
(14) Ryhl-Svendsen, M.; Glastrup, J. Atmos. Environ. 2 0 0 2 , 36, 3909-3916.
(15) Kobayashi, K.; Tanaka, M.; Kawai, S. J. Chromatogr. 1 9 8 0 , 187, 413-417.
(16) Magin, D. F. J. Chromatogr. 1 9 7 9 , 178, 219-227.
(17) Swarin, S. J.; Lipari, F. J. Liq. Chromatogr. 1 9 8 3 , 6, 425-444.
(18) Schmied, W.; Przewosnik, M.; Baechmann, K. Fresenius’ Z. Anal. Chem.
1 9 8 9 , 335, 464-468.
* Corresponding author. Fax: +81-3-3700-9298. E-mail: uchiyama@nihs.go.jp.
† National Institute of Health Sciences.
‡ Chiba University.
§ Present address: Faculty of Pharmacy, Musashino University, 1-1-20,
Shinmachi, Nishi-Tokyo City, Tokyo, 202-8585, Japan.
(1) Mader, P. P.; Cann, G.; Palmer, L. Plant Physiol. 1 9 5 5 , 30, 318-323.
10.1021/ac0493471 CCC: $27.50 © 2004 American Chemical Society
Published on Web 08/27/2004
Analytical Chemistry, Vol. 76, No. 19, October 1, 2004 5849

pentafluorophenylhydrazine,20 and 2,4-dinitrophenylhydrazine
(DNPH).21-23 Indeed, for over 70 years, aldehydes and ketones
have been characterized through their reaction with DNPH to
the corresponding hydrazone derivatives.24,25 The derivatives are
then separated using HPLC with UV detection at a wavelength
near the absorption maximums of the respective hydrazones
(typically 360 nm).26,27 In recent years, the DNPH-HPLC method
has been established as the most widely used standard procedure
for the determination of aldehydes and ketones in air samples.
In solid and liquid samples, such as foods, beverages, and
biological materials, 2-nitrophenylhydrazine is widely used for the
derivatization of carboxylic acids, aldehydes, and ketones;28,29
however, there is no report describing simultaneous analysis of
aliphatic carboxylic acids, aldehydes, and ketones in an air sample.
It is advantageous to measure carboxylic acid and aldehyde levels
simultaneously because they are sometimes emitted from the
same sources in air. In this study, we developed a new method
for simultaneous measurement of C1-C4 aliphatic carboxylic acids
and aldehydes using DNPH derivatization.
acid 99%, formaldehyde (37% solution in water), acetaldehyde
99.5%, propionaldehyde 97%, butyraldehyde 99.5%, and silica gel
(Davisil grade 635, 60-100 mesh) were from Aldrich (Milwaukee,
WI). DNPH (containing ∼50% water) and phosphoric acid (>85%)
were pure grade from Wako Pure Chemical Industries Ltd. 2,4-
Dinitrophenylhydrazine hydrochloride (>85%) was from Kanto
Chemical Co. (Tokyo, Japan).
Synthesis of Carboxylic 2 ,4 -Dinitrophenylhydrazide De-
rivatives. Formic 2,4-Dinitrophenylhydrazide (Formic DNPhy-
drazide). Five grams of 2,4-dinitrophenylhydrazine hydrochloride
was dissolved in 50 mL of formic acid. Pure water (50 mL) was
added to this solution, and a light yellow precipitate formed. After
2 h, the precipitate was recovered by filtration and washed three
times with 500 mL of water. The precipitate of the hydrazide
derivative was recrystallized from ethanol and vacuum-dried.
Acetic 2,4-Dinitrophenylhydrazide (Acetic DNPhydrazide), Pro-
pionic 2,4-Dinitrophenylhydrazide (Propionic DNPhydrazide), and
Butyric 2,4-Dinitrophenylhydrazide (Butyric DNPhydrazide). Five
grams of 2,4-dinitrophenylhydrazine hydrochloride was dissolved
in 50 mL carboxylic acid with 5 mL concentrated sulfuric acid.
After 24 h at room temperature, the precipitate was recovered by
filtration and washed three times with 500 mL of water. The
precipitates of the hydrazide derivatives were recrystallized from
ethanol and vacuum-dried.
P reparation of the DNP H Cartridge for Collection of
Carboxylic Acids. Fifty grams of silica gel was washed three
times with 500 mL of water; two times with 500 mL of methanol;
and last, two times with 500 mL of acetonitrile. 2,4-Dinitrophen-
ylhydrazine hydrochloride (0.18 g) and 0.5 mL of phosphoric acid
were dissolved in 150 mL of acetonitrile. This solution was added
to 50 g of washed silica gel. The mixture was stirred, and then
the solvent was evaporated to dryness at 40 °C under vacuum on
a rotary evaporator. The DNPH-coated silica particles (500 mg)
were packed into a polyethylene cartridge (50 mm long × 8.8
mm i.d.) and stored in a refrigerator (4 °C).
EXPERIMENTAL SECTION
Apparatus. The HPLC system (Shimadzu, Kyoto, Japan) used
included two LC-10ADvp pumps, an SIL-10ADvp autosampler, an
SPD-10Avp ultraviolet absorbance detector adjusted to 350 nm,
and an SPD-M10Avp photodiode array detector. The analytical
column was a 250 mm long × 4.6 mm i.d. stainless steel tube
packed with RP-Amide C16, 5-µm particle size (Supelco Inc,
Bellefonte, PA). Solution A of the mobile phase mixture was
acetonitrile/ water, 40/ 60 v/ v, and solution B was 60/ 40 v/ v. HPLC
was carried out with 100% A for 8 min, followed by a linear gradient
from 100% A to 100% B in 22 min and then held for 15 min. The
flow rate of the mobile phase was 1.5 mL/ min. A Discovery C18
column, 5-µm particle size 250 mm long × 4.6 mm i.d. (Supelco
Inc, Bellefonte, PA) was used as a reference, and the RP-Amide
C16 column was used for most experiments.
Melting points of the DNPH derivatives were determined by
a Q10 differential scanning calorimeter (DSC, TA Instruments,
New Castle, DE) under nitrogen atmosphere.
The carboxylic acid DNPH derivatives were characterized by
NMR spectroscopy using a JNM-ECX400 spectrometer (JEOL
Ltd., Tokyo, Japan) operating at 400 MHz.
Reagents. The water used in HPLC and sample preparation
was deionized and purified using a Milli-Q Water System equipped
with a UV lamp (Millipore, Bedford, MA). Acetonitrile was HPLC
grade from Wako Pure Chemical Industries Ltd. (Osaka, Japan).
Formic acid 96%, acetic acid 99.8%, propionic acid 99.5%, butyric
Generation of Standard Formic Acid Gas and Collection
using DNP H Cartridge. Standard gas was generated in a
dynamic system by diffusing formic acid using a gas generator
(Permeater PD-1B, GASTEC Co., Kanagawa, Japan). Pure, pres-
sure-regulated air was passed with a constant flow rate through
the PD-1B gas generator including a dispersion bottle (3 mm i.d.)
containing ∼ 2 mL of formic acid and thermostated at 30.0 °C.
The concentration of generated gas was calculated from the weight
loss (mg) of the dispersion bottle and the pure air volume (L)
passed into the PD-1B; e.g. 5 ppm of standard formic acid gas is
obtained from a flow rate of 1.35 L/ min at 30 °C. One end of the
DNPH cartridge was attached to the testing chamber, and the
other end was connected to a sampling pump equipped with a
mass flow controller (SP 208 Dual, GL Sciences Inc., Tokyo,
Japan).
(19) Bueldt, A.; Lindahl, R.; Levin, J. O.; Karst, U. J. Environ. Monit. 1 9 9 9 , 1,
39-43.
(20) Cecinato, A.; Di-Palo, V.; Mabilia, R.; Possanzini, M. Chromatographia 2 0 0 1 ,
54, 263-269.
(21) Grosjean, D. Environ. Sci. Technol. 1 9 8 2 , 16, 254-262.
(22) Kuwata, K.; Uebori, M.; Yamasaki, H.; Kuge, Y.; Kiso, Y. Anal. Chem. 1 9 8 3 ,
55, 2013-2016.
RESULTS AND DISCUSSION
(23) Levin, J. O.; Andersson, K.; Lindahl, R.; Nilsson, C. A. Anal. Chem. 1 9 8 5 ,
Carboxylic DNP hydrazide Derivatives. A DNPH cartridge
57, 1032-1035.
saturated with formic acid vapor becomes gradually discolored
and completely changes to light yellow in 6 h. The HPLC
chromatogram of the eluant from this DNPH cartridge indicates
complete consumption of DNPH accompanied by formation of
formic 2,4-dinitrophenylhydrazide (formic DNPhydrazide). Figure
1 shows the peak area changes with time of DNPH and formic
(24) Allen, C. F. H. J. Am. Chem. Soc. 1 9 3 0 , 52, 2955-2959.
(25) Brady, O. L. J. Chem. Soc. 1 9 3 1 , 756-759.
(26) Bohlmann, F. Chem. Ber. 1 9 5 1 , 84, 490-504.
(27) Rappoport, Z.; Sheradsky, T. J. Chem. Soc. 1 9 6 8 , 277-291.
(28) Peters, R.; Hellenbrand, J.; Mengerink, Y.; Van der Wal, S. j. J. Chromatogr.,
A 2 0 0 4 , 1031, 35-50.
(29) Miwa, H. J. Chromatogr., A 2 0 0 0 , 881, 365-385.
5850 Analytical Chemistry, Vol. 76, No. 19, October 1, 2004

Table 1. Maximum Absorption Wavelengths (λMax) and
Molar Absorption Coefficients (E) of C1-C4 Carboxylic
Acid and Aldehyde DNPH Derivativesa
λmax
ꢀ
mp
(°C)
(nm) (L/ mol/ cm)
formic DNPhydrazide
acetic DNPhydrazide
propionic DNPhydrazide
butyric DNPhydrazide
DNPhydrazine
formaldehyde DNPhydrazone
acetaldehyde DNPhydrazone
propionaldehyde DNPhydrazone 360
331
333
334
334
351
349
359
1.36 × 104
1.42 × 104
1.44 × 104
1.44 × 104
1.91 × 104
2.10 × 104
2.10 × 104
2.18 × 104
185.7
201.5
192.0
167.1
1.55 × 104 ∼20030
153-15631
165-16831
152-15531
117-11931
butyraldehyde DNPhydrazone
361
a
Melting points of aldehyde-2,4-DNPhydrazones are the literature
values.
Figure 1. The reaction of adsorbed formic acid and DNPH with
time. (λ ) 360 nm).
samples prepared by direct reaction of the carboxylic acids with
DNPH (Experimental Section). Two HPLC columns with differing
selectivities (C18 and RP-Amide C16) were used to confirm the
retention time assignments.
DNPhydrazide at wavelength 360 nm. Acetic acid, propionic acid,
and butyric acid exhibit similar behavior with longer reaction time
in order of increasing carbon number.
Characterization. The structures of the synthesized hy-
1
drazides are consistent with the NMR data. H NMR data of the
derivatives were as follows: formic DNPhydrazide 1H NMR (CD3-
CN, TMS), δ (ppm): 7.25 (d 1H, J ) 9.6, a), 8.28 (dd 1H, J ) 2.5,
9.4, b), 8.94 (d 1H, J ) 2.8, c), 9.36 (s 1H, d), 8.49 (s 1H, e), 8.23
(s 1H, f). Acetic DNPhydrazide 1H NMR (CD3CN, TMS), δ
(ppm): 7.25(d 1H, J ) 9.2, a), 8.25 (dd 1H, J ) 2.75, 9.6, b), 8.92
(d 1H, J ) 2.8, c), 9.37 (s 1H, d), 8.49 (s 1H, e), 2.02 (s, 3H, CH3).
Propionic DNPhydrazide 1H NMR (CD3CN, TMS), δ (ppm): 7.22
(d 1H, J ) 9.6, a), 8.23 (dd 1H, J ) 2.52, 9.39, b), 8.91 (d 1H, J )
2.29, c), 9.37 (s 1H, d), 8.48 (s 1H, e), 2.30 (t, 2H, CH2), 1.13 (t,
3H, CH3). Butyric DNPhydrazide 1H NMR (CD3CN, TMS), δ
(ppm): 7.21 (d 1H, J ) 9.15, a), 8.24 (dd 1H, J ) 2.75, 9.62, b),
8.91 (d 1H, J ) 2.75, c), 9.39 (s 1H, d), 8.49 (s 1H, e), 2.27 (m,
2H, J ) 7.56, CH2), 1.66 (t, 2H, J ) 7.42, CH2), 0.964 (t, 3H, J )
7.327, CH3).
It is postulated that carboxylic acids react with DNPH to
initially form corresponding hydrazone derivatives, which then
isomerize to hydrazides by keto-enol tautomerization.
Carboxylic DNPhydrazide derivatives demonstrated excellent
thermal stability on DSC analysis. Melting points are listed in
Table 1 and are higher than those of corresponding hydrazones
by 32-50 °C.
UV-visible absorption spectra of the carboxylic DNPhydrazide
derivatives (20 µmol/ L acetonitrile solution) are presented in
Figure 2. For reference, the absorption spectrum of DNPH is also
shown.
The spectral profiles of carboxylic DNPhydrazide derivatives
exhibited similar curves. Specifically, acetic, propionic, and butyric
DNPhydrazide showed much the same maximum absorption
wavelengths (333-334 nm) and molar absorption coefficients (1.42
× 104 to 1.44 × 104 L/ mol/ cm) (see Table 1).
Separation of Carboxylic DNP hydrazide Derivatives. Ana-
lytical conditions for formic, acetic, propionic, and butyric DN-
Phydrazide derivatives were examined using a C18 and an RP-
Amide C16 column. Figure 3 shows a chromatogram of C1-C4
(30) Budavari, S.; O’Neil, M.; Smith, A.; Heckelman, P.; Kinneary, J. The Merck
Index. An Encyclopedia of Chemicals, Drugs and Biologicals; Chapman and
Hall: London, UK, 1996; p 2576.
(31) The 2003-2004 Aldrich Handbook of Fine Chemicals; Sigma-Aldrich: St.
Louis, MO.
The derivatives eluted from the DNPH cartridges were identi-
fied by comparison of the HPLC retention times with authentic
Analytical Chemistry, Vol. 76, No. 19, October 1, 2004 5851

Figure 2. UV-visible absorption spectra of carboxylic DNPhydrazide
derivatives in acetonitrile solution (20 µmol/L).
Figure 4. Reaction of formic acid and DNPH in the DNPH cartridge
at various temperatures.
DNPH peak, and the butyric DNPhydrazide peak was detected
after the DNPH peak. In the case of the C18 column, the formic
DNPhydrazide peak overlapped with the acetic DNPhydrazide
peak. In addition, the Z isomer of the acetaldehyde DNPhydrazone
peak overlapped with the E isomer. The RP-Amide C16 column
gave completely separated peaks. The amide column is an
alkylamide reversed-phase column based on high-purity silica. The
presence of the alkylamide group gives these columns an
alternative selectivity compared to the simple hydrophobic selec-
tivity of the C18 columns.35 Due to the presence of both alkyl and
amide functionality on the RP-Amide C16 column, two retention
mechanisms are available: a mechanism based on the hydropho-
bic alkyl groups and a mechanism based on interaction between
the polar groups of the carboxylic DNPhydrazide derivatives and
the amide group. Therefore, amide columns, such as RP-Amide
C16 or Bonus-RP, provide unique selectivity that leads to sufficient
resolution between acetic and formic DNPhydrazide derivatives.
In the chromatograms produced using the alkylamide column,
C1-C4 straight-chain carboxylic DNPhydrazides and aldehyde
DNPhydrazones were completely separated.
The limits of detection (LOD) by HPLC, defined as three times
the standard deviation of 10 replicate measurements at 10 µg/ L,
were as follows: formic acid 1.0, acetic acid 0.7, propionic acid
1.9, butyric acid 2.4, formaldehyde 1.0, acetaldehyde 1.4, propi-
onaldehyde 1.3, and butyraldehyde 1.9 µg/ L.
Collection and Derivatization of Formic Acid Using the
DNP H Cartridge. Two DNPH cartridges (primary and second-
ary) were connected in series, and standard formic acid gas (5
ppm) was drawn through the cartridges for 30 min. The secondary
cartridge served to collect any material that may have escaped
the primary cartridge. Immediately after collection, the DNPH
cartridges were heated for 1-10 h in a thermostat. Later, the
DNPH derivatives were removed from the adsorbent by eluting
acetonitrile (5 mL) through the DNPH cartridge via a syringe to
a graduated test tube over a 1-min time period. The eluate was
then analyzed by HPLC. Figure 4 shows the changes in formic
acid derivative content with temperature. Under all conditions
Figure 3. Chromatographic profile of C1-C4 carboxylic DNPhy-
drazides and aldehyde DNPhydrazones on an RP-Amide C16 column
(100 µmol/L). A prime sign indicates the Z isomer peak; the other
peaks without a prime sign are the E isomers.32 FC, formic DNPhy-
drazide; AC, acetic DNPhydrazide; PC, propionic DNPhydrazide; BC,
butyric DNPhydrazide; FA, formaldehyde-DNPhydrazone; AA, ac-
etaldehyde DNPhydrazone; PA, propionaldehyde DNPhydrazone; BA,
butyraldehyde DNPhydrazone.
hydrazide and hydrazone derivatives using an RP-Amide C16
column.
The reference solution was 100 µmol/ L and contained 0.1%
phosphoric acid. Some small peaks, corresponding to the Z
isomers, appeared in the chromatograms.32 Aldehyde-DNPhydra-
zones in the standard solution (non-acid) exhibits only the E
isomer; however, this E isomer can partly convert into the Z
isomer upon addition of acid solution,33,34 and the sample solution
extracted from the DNPH cartridge inevitably contains phosphoric
acid. The isomerization involves addition of a nucleophile (phos-
phate ion or water) to the protonated E or Z isomers, and an
equilibrium Z/ E isomer ratio was observed in 0.02-0.2% v/ v
phosphoric acid solutions.32
In Figure 3, the chromatographic peaks of formic, acetic, and
propionic DNPhydrazide derivatives were detected before the
(32) Uchiyama, S.; Ando, M.; Aoyagi, S. J. Chromatogr., A 2 0 0 3 , 996, 95-102.
(33) Behforouz, M.; Bolan, J. L.; Flynt, M. S. J. Org. Chem. 1 9 8 5 , 50, 1186-
1189.
(34) Binding, N.; Mu¨ ller, W.; Witting, U. Fresenius’ J. Anal. Chem. 1 9 9 6 , 356,
315-319.
(35) Kallury, K.; Shieh, P.; Paschal, R.; Cooke, N. Supelco Rep. 1 9 9 8 , 17, 5-7.
5852 Analytical Chemistry, Vol. 76, No. 19, October 1, 2004

Table 2. Blank Value of the DNPH Cartridge
20 °C
80 °C
mean
(µg)
SD
(µg)
CV mean
SD
(µg)
CV
(%)
(%)
(µg)
acetic acid
formic acid
0.0041 0.0012 29
3.1
0.095
0.020
0.048
nda
3.0
4.3
7.3
0.029
0.028
nda
0.0028
0.0027
nda
9.5 0.48
9.5 0.66
nda
propionic acid
butyric acid
formaldehyde
acetaldehyde
0.019
0.0009
4.8 0.31
0.013
4.5
7.4
nda
nda
0.077 0.0057
0.017 0.0032 18
0.050 0.0042 8.2
prorionaldehyde nda
nda
butyraldehyde
nda
nda
a
Not detected.
measurement of carboxylic acids and aldehydes is 1% v/ w
phosphoric acid because acidic conditions are needed for the
measurement of aldehydes.
On the basis of these results, the following optimized condi-
tions for derivatization were identified: DNPH cartridge with 1%
v/ w phosphoric acid, 80 °C reaction temperature, and 5 h reaction
time.
Blank Test and LOD. Seven DNPH cartridges were kept for
5 h in a thermostat at 80 or 20 °C and analyzed by the above-
mentioned procedure. The blank values of C1-C4 carboxylic acids
and aldehydes are shown in Table 2.
The blank values in the case of heating at 80 °C were higher
than at 20 °C, especially for carboxylic acids. It is believed that
the unreacted background compounds in the DNPH cartridge
formed DNPH derivatives on heating. The LOD for measurement
of formic acid, defined as three times the standard deviation of
the blank (3σ), is 0.06 µg per cartridge. When the collection of
air sample occurs for 24 h at 100 mL/ min (144 L), the LOD is 0.4
µg/ m3.
Measurement of Indoor Air. Indoor air in an apartment
house was passed through the DNPH cartridges for 24 h at a
flow rate of 50 mL/ min. After sampling, the cartridges were
maintained for 5 h in a thermostat at 80 or 20 °C and then analyzed
by the above-mentioned procedure. The chromatograms in Figure
6 illustrate measurements made under the two different temper-
ature conditions.
It was confirmed that carboxylic acids and aldehydes in sample
air were completely collected in the primary DNPH cartridge
because the amounts of compounds detected in secondary
cartridges were equivalent to background levels. The concentra-
tions of aldehydes including formaldehyde, acetaldehyde, and
propionaldehyde following heating at 80 °C were the same as
those at 20 °C. Formic acid collected with the DNPH cartridge
was detected following heating at 80 °C.
Some unknown peaks were detected in Figure 6. In samples
without heat treatment, peaks of “unknown 1” were often detected
in the sample solution immediately after collection of the air
sample. The peak corresponding to “unknown 2 ” was detected
in the sample solution from the DNPH cartridges stored for longer
time periods. This suggests that unknown 1 was converted to
unknown 2 . The retention time for unknown 1 fits well to 2,4-
dinitrophenyl azide, a known interfering agent of the DNPH
method resulting from the reaction of the reagent with nitrogen
dioxide.36 This compound is known to be thermolabile. It decom-
Figure 5. Influence of phosphoric acid concentration on formation
of hydrazide derivative.
studied, no formic acid derivative was detected in the secondary
cartridge, indicating that no “breakthrough” of the primary
cartridge occurred.
Formic acid was physically adsorbed onto the DNPH cartridge
at a rapid rate and reacted chemically with DNPH at a slow rate.
The rate of reaction was very slow at 20 °C and increased
dramatically with temperature. The reaction rate at 100 °C was
the most rapid of all conditions tested; however, recovery did not
reach 100% (0.3 µmol) and began to decrease after 2 h. Thermal
decomposition of DNPH was observed. During heating at 100 °C,
the peak area of DNPH decreased linearly to 14% in 10 h, and
decomposition products (1,3-dinitrobenzene and 2,4-dinitroaniline)
were detected as overlapping peaks at 6.5 min on the chromato-
gram. It is believed that the decomposition of DNPH induced a
poor recovery. During heating at 80 °C, the peak area of DNPH
decreased to 76% in 10 h. However, at 20-60 °C, there were no
decreases of DNPH. From these results, the standard heating
condition for derivatization to hydrazide was selected as 80 °C
and 5 h.
Carboxylic acids and aldehydes react with DNPH in the
presence of an acidic catalyst; therefore, the effect of the amount
of phosphoric acid was examined for the DNPH cartridge. DNPH-
coated silica gels containing 0.1-10% v/ w phosphoric acid (as 85%
phosphoric acid) were prepared, and 500 mg was packed into
polyethylene cartridges. All cartridges were connected to a
secondary cartridge containing 1% v/ w phosphoric acid, and
standard formic acid gas (5 ppm) was drawn through for 30 min.
Immediately after collection, DNPH cartridges were heated for 5
h at 80 °C. Figure 5 shows the recovery rates as a function of
phosphoric acid concentration.
It appears that formic acid is difficult to derivatize under strong
or weakly acidic conditions. However, complete adsorption of
formic acid is believed to have occurred in all primary DNPH
cartridges containing 0.1-10% v/ w phosphoric acid since no
formic acid was detected in any secondary cartridges. Therefore,
the recovery rate in Figure 5 is not a reflection of the collection
efficiency. The derivatization reaction to hydrazide progressed
essentially to completion for the DNPH cartridges containing 0.2-
1% v/ w phosphoric acid. The best condition for simultaneous
Analytical Chemistry, Vol. 76, No. 19, October 1, 2004 5853

Figure 6. Chromatographic profiles of DNPH derivatives from air sample measured with a heat treatment of 80 and 20 °C. FC, formic
DNPhydrazide; AC, acetic DNPhydrazide; PC, propionic DNPhydrazide; FA, formaldehyde DNPhydrazone; AA, acetaldehyde DNPhydrazone;
PA, propionaldehyde DNPhydrazone; BA, butyraldehyde DNPhydrazone; AN, acetone; DB, 1,3-dinitrobenzene; DA, 2,4-dinitroaniline.
poses under thermal stress to nitrobenzofurazan-3-oxide (unknown
permit the simultaneous measurement of carboxylic acids and
aldehydes in air using a DNPH cartridge containing DNPH-coated
silica gel. Heat treatment of the collected DNPH cartridge was
necessary in order to measure carboxylic acid levels because the
reaction times of derivatization are very slow. In the case of formic
acid, 5 h reaction time at 80 °C is needed after collection.
2 ), which is more stable.37,38
CONCLUSIONS
In this study, it was found that C1-C4 carboxylic acids reacted
with DNPH to form the corresponding hydrazide derivatives.
These compounds are thermally stable and exhibit maximum
absorption wavelengths near the aldehyde DNPhydrazone deriva-
tives. C1-C4 carboxylic DNPhydrazides and aldehyde DNPhy-
drazones were completely separated and detected by HPLC using
a RP-Amide C16 column and 350-nm detection. These properties
ACKNOWLEDGMENT
This work was supported by grants from the Health and Labor
Sciences Research from the Ministry of Health, Labor and Welfare
of the Japanese Government.
(36) Karst, U.; Binding, N.; Cammann, K.; Witting, U. Fresenius’ J. Anal. Chem.
1 9 9 3 , 345, 48-52.
(37) Dyall, L. K.; Kemp, J. E. J. Chem. Soc. B 1 9 6 8 , 976-979.
(38) Dyall, L. K.; Wah, W. M. Aust. J. Chem. 1 9 8 5 , 38, 1045-1059.
Received for review May 3, 2004. Accepted July 21, 2004.
AC0493471
5854 Analytical Chemistry, Vol. 76, No. 19, October 1, 2004