A new fluorimetric method for the determination of formaldehyde in air based on the liquid droplet sampling technique

Abstract

A new, simple, sensitive, selective and in-field fluorimetric method for the determination of formaldehyde is proposed. The reaction of formaldehyde with hydralazine in acidic medium, heating on a boiling water-bath for 25 min, produces s-triazolo[3,4-a]phthalazine (Tri-P). The fluorescence intensity of the product formed (Tri-P) was determined at λ_em = 389 nm with λ_ex = 236 nm. The fluorescence intensity is linear over a formaldehyde concentration range of 1.2–33.0 μg l⁻¹. The proposed method was applied successfully to the determination of formaldehyde sampled from the atmosphere using the liquid droplet technique. Formaldehyde vapour in a wind tunnel was produced by a mean of permeater. A linear curve was obtained between the concentration in the wind tunnel and that in the droplet. The detection limit for formaldehyde was 2.0 μg l⁻¹ with RSDs varying from 3 to 12% in ambient air, using a droplet correction solution (boric acid and hydralazine). The effect of interfering substances on the determination shows that most cations and anions did not interfere. The results obtained were satisfactory compared with a reference method.

---

1 Introduction

Formaldehyde is used as a coating, resin and/or adhesive in many building materials. There are several types of formaldehyde resin mixtures. Urea–formaldehyde contributes the most to indoor air pollution, because of its water solubility. Diverse health effects have been attributed to formaldehyde (CH₂O) exposure in residential and occupational environments. The main indoor sources of CH₂O are wood products, such as medium density fibreboard (MDF) and particleboard. MDF is used in the manufacture of furniture, cabinets and shelving. The CH₂O mixture used in building materials releases CH₂O vapour into the air. The rate at which the vapour is released depends on the air temperature and relative humidity. An increase in temperature of 5–6 °C can double the concentration, and an increase in relative humidity from 30 to 70% can cause a 40% rise in CH₂O concentration. If both the temperature and relative humidity are raised, the concentration can increase to as much as five times its original level. Formaldehyde is also known to be an irritant to the eyes, nose and throat,¹ and it has been declared a suspected carcinogen.² Therefore, because of its toxicity and possible carcinogenic properties, many countries have established an 8 h time average permissible exposure limit for CH₂O ranging from 0.6 to 2.5 mg L⁻¹.³ Moreover, CH₂O is now recognized as one of the most important indoor air pollutants,⁴ and since people spend about 90% of their time indoors,⁵ it is important to determine the indoor concentration levels to obtain an estimate of personal exposure. Hence it is of great concern to develop a rapid and sensitive method for monitoring formaldehyde in indoor air.

Spectrophotometric methods are the most widely used approach.^6–8 However, these methods suffer from a number of interferences and the level of detection is insufficient to monitor occupational exposure. Gas chromatographic techniques for the determination of formaldehyde in water have been reported.⁹ Various devices have been developed for sampling formaldehyde from ambient air.^10–12 Most of these methods are not convenient in field investigations and especially are not suitable for monitoring personal exposure.

This paper describes the fluorimetric determination of low levels of formaldehyde in indoor air using hydralazine as a fluorescent reagent. The utilization of hydralazine in a spectrofluorimetric method for the determination of formaldehyde has not been reported previously. The sensitivity and the limit of detection were improved compared with other methods.^8,9 The droplet method is one of the gas passive sampling monitoring procedures. A liquid droplet containing trapping solution is formed at the end of a stainless steel pipe. Formaldehyde vapour in ambient air diffuses to a growing droplet solution and is dissolved in it. After a gas-trapping process, the gas droplet is collected in a vial and analysed by spectrofluorimetry. This paper describes the parameters such as diameter (od = 40 mm and id = 25 mm), size of the pipe, flow rate of the reagent solution and boiling time that are needed for the reaction between formaldehyde and hydralazine. The droplet method was applied to monitor formaldehyde in real indoor samples with satisfactory results.

2 Experimental

2.1 Apparatus

An RF-1500 spectrofluorimeter (Shimadzu, Japan) was used for recording the fluorescence intensity. A controlled water-bath was used to maintain the temperature. A 1 × 1 cm quartz cell was used for recording spectra and making fluorescence measurements.

2.2 Reagents

All chemicals were of analytical-reagent grade and de-ionized (DI) and doubly distilled (DD) water was used throughout.

Hydralazine solution. Hydralazine (Wako) was used without further purification. Hydralazine was dissolved at 0.100 mM in 0.100 M HCl. Boric acid (0.080 M) was dissolved in DI-DD water.

Formaldehyde stock standard solution (10.0 mM). This was prepared in DI-DD water. Care must be taken while preparing formaldehyde. Working standard solutions were prepared by further dilution just before use.

2.3 Calibration

Into a series of boiling test-tubes, a known concentration of formaldehyde standard solution was added, followed by 0.040 ml of hydralazine solution. The test-tubes were then placed in a boiling water-bath for 25 min for completion of the reaction. The test-tubes were then cooled under tap water and transferred into a 10 ml calibrated flask and diluted with DI-DD water. The fluorescence intensity was measured at λ_ex = 236 and λ_em = 389 nm. The fluorescence intensity was valid over a formaldehyde concentration range of 1.2–33.0 μg l⁻¹.

Standard concentrations of gaseous formaldehyde in dried and filtered compressed air were generated by means of a permeater having a constant rate of formaldehyde release. The sampling time was 60 min. The flow rate of the formaldehyde gas was varied from 2.8 to 6.2 l min⁻¹ to adjust the air formaldehyde concentrations. The air stream containing the formaldehyde was equilibrated thermally with glass beads at ambient laboratory temperature, as measured by thermometer. The gaseous formaldehyde passed through the tube of the D-10, and the incubation gas tank temperature was 35 °C. Formaldehyde gas was absorbed from the air stream in 10–15 ml of (1) 0.080 M boric acid and (2) 0.080 M boric acid + 0.100 mM hydralazine (1 + 1). The trapped solution was pumped by means of a PC-5000 pump (Sanuki-koug you, Japan) for droplet making. The formaldehyde gas stream flowed inside a wind tunnel made from an acrylic cylinder. The tunnel was used to control and keep the formaldehyde gas concentration produced by the permeater. The concentration of formaldehyde generated by the permeater was calculated by the equation C = KD × 1000F⁻¹, where C = CH₂O gas concentration (v/v), D = diffusion velocity (0.38 μg min⁻¹), F = calibrated flow rate of diluted gas mass (l min⁻¹) and K = CH₂O gas mass converted to gas volume (K = 0.816) . The droplet device was placed inside the wind tunnel and the formaldehyde gas was trapped by the trapping solution. The sampling device for making and collecting formaldehyde is shown in Fig. 1. Samples were analyzed immediately (within 30 min) to reduce the loss of formaldehyde prior to the analysis procedure.

Fig. 1 Sampling device for the formation of the droplet and for collecting formaldehyde.

Samples of each of six air concentrations were collected in sets of five. Each sample in each set was analysed in order to establish the precision of the sample collected. The actual air concentration of formaldehyde was calculated on the basis of the known permeation rate of formaldehyde release.

2.4 Procedure for formaldehyde sampled from air by the means of liquid droplet method

Two procedures were adopted for the absorption of formaldehyde from air. Procedure (1): boric acid solution (0.080 M) is used as a trapping solution; 15 ml were placed in a test-tube and pumped by a means of PC-5000 pump (Sanuki-koug you, Japan) with a flow rate controlled at 0.010 l min⁻¹. Boric acid was passed through a stainless steel capillary and a droplet was formed at the tip of the capillary and collected for analysis for formaldehyde absorbed from air. The formaldehyde gas concentration in the wind tunnel was adjusted by a mean of permeater (Castec PD-1B). The standard formaldehyde gas in the wind tunnel was forcibly circulated with a fan to ensure an equal distribution of the CH₂O gas inside the wind tunnel (Fig. 2). The droplet formed was collected and then treated as described in the procedure for the formaldehyde calibration curve. Procedure (2): boric acid (0.080 M)–hydralazine (0.100 mM) was used. The same procedures were followed as described for procedure (1). The mixed solution of boric acid and hydralazine was used for droplet correction for the absorption of CH₂O from air.

Fig. 2 Schematic diagram of the wind tunnel and the droplet device.

3 Results and discussion

3.1 Optimization of fluorescence reaction

Analytical procedure. A 0.1 ml volume of the collected droplet, which contained formaldehyde (absorbed by both of the trapping solutions), was placed in a test-tube, followed by the addition of 0.040 ml of hydralazine. The test-tube was placed in a boiling water-bath for 25 min, cooled and transferred carefully into a 5 ml calibrated flask and diluted to volume with water. The concentration of formaldehyde in air collected by the droplet method was calculated from the following equation: where X = concentration of formaldehyde (ppb); F = fluorescence intensity, C = intercept and B = slope of the regression equation of the calibration curve, i.e., A = 10.1X + 12.4, and C′ and B′ are the intercept and the slope of the regression equation obtained between the concentration of formaldehyde in the wind tunnel and the droplet, i.e., A = 10.3X −68.6.

Reaction product. The reaction of hydralazine with formaldehyde in acidic medium leads to the formation of s-triazolo[3,4-a]phthalazine (Tri-P). The reaction product is well established^13–16 and shown in Scheme 1.

Scheme 1 Production of Tri-P from reaction of hydralazine with formaldehyde.

Fluorescence spectra. The formation of Tri-P under the optimum experimental conditions, gave a highly fluorescent compound that exhibits an excitation wavelength at 236 nm and emission wavelength at 389 nm. The fluorescence spectra are shown in Fig. 3.

Fig. 3 Fluorescence intensity of the reaction product Tri-P. Hydralazine, 0.040 ml; formaldehyde, 6.0 μg ml⁻¹; λ_ex = 236 nm and λ_em = 389 nm.

Effect of acid and hydralazine concentration on the formation of Tri-P. When hydralazine solution is dissolved in hydrochloric acid, a highly fluorescent compound is produced. This shows that the formation of Tri-P depends on the acid, hence 0.1 M HCl was used throughout the experiment. Hydralazine concentrations in the range 0.0001–0.001 mM were investigated. The fluorescence intensity was stable from 0.0003 to 0.0006 mM hydralazine. Therefore, 0.04 ml of 0.1 mM hydralazine was adopted.

Effect of temperature. The effect of temperature on the fluorescence intensity was checked in the range 50–100 °C. It was observed that 100 °C give the highest fluorescence intensity and this was adopted in subsequent work.

Effect of reaction time. The formation of Tri-P was completed within 25 min on a boiling water-bath. The fluorescence intensity gradually increased with time and remained constant after 25 min.

Calibration curve. According to the proposed method, the fluorescence intensity was proportional to formaldehyde concentration in the range 1.2–33 ppb. The linear regression equation was A= 10.2X + 12.4, with a correlation coefficient r² = 0.9988. The standard deviation of the calibration curves and the relative error of the intercept and the slope were found to be 4.7, 3.8 and 0.57, respectively.

The accuracy of the proposed method was tested by the analysis of standard formaldehyde solution over a period of 15 days (n = 5). The RSD for formaldehyde concentrations of 5.0, 15 and 25 ppb were 0.65, 0.43 and 0.58%, respectively.

3.2 Optimization of liquid droplet sampling

Effect of stainless steel valve diameter on droplet formation. Different outer diameters of the stainless steel valve were examined. It was found that an od of 5.0 mm gave better results than with 1.6 and 4.0 mm od. The results are summarized in Table 1. Moreover, as the stainless steel diameter increased and the wind velocity increased, the collection of a higher concentration of CH₂O could be achieved. The outer diameter of the valve is an important factor and the droplet volume is linearly related to the valve diameter. This is in agreement with Tate’s law.¹⁷

Table 1 Effect of stainless steel pipe diameter on the droplet (n = 5)

Stainless steel od/mm
Parameter	1.6	4.0	5.0
Trapping reagent volume/ml	0.024 ± 0.003 (13%)	0.039 ± 0.003 (8%)	0.062 ± 0.002 (3%)
Trapping time/s	162 ± 19.0 (12%)	265 ± 27.0 (10%)	448 ± 13.0 (3%)
CH2O concentration in droplet (ppb)	60 ± 0.0	50 ± 5.0	70 ± 0.0

---

Effect of flow rate on absorption of CH₂O gas by the droplet. The collection efficiency of CH₂O using the liquid droplet method was high. Owing to the higher collection efficiency of the gas phase when using droplet method, the flow rate of the absorbent solution (i.e., boric acid and hydralazine) was checked at 0.01, 0.02 and 0.03 ml min⁻¹. Fig. 4 shows the effect of the sample flow rate on the collection of the droplet containing CH₂O. A decrease in CH₂O concentration in the droplet was observed, when the flow rate increased. A flow rate of 0.01 ml min⁻¹ was adopted throughout the experiment, since it gave the highest CH₂O concentration in the droplet. A decrease in CH₂O concentration was observed when the flow rate increased. This may be attributed to the fact that less CH₂O was collected and hence had an adverse effect on the efficiency of the chromogen and the signal. Therefore, a change in the signal was observed when the flow rate changed. A correlation was obtained between CH₂O concentration in the wind tunnel and the droplet.

Fig. 4 Effect of flow rate on the concentration of formaldehyde in the droplet collected from air.

3.3 Determination of formaldehyde in air

Determination of formaldehyde gas sampled from the atmosphere by the proposed method. Indoor exposure experiments were performed on-site in a household and a cupboard. In each case, triplicate sets of sampling devices for droplets were exposed according to the recommended procedures. The results are presented in Table 2 and show good agreement between the droplet method and a reference method.⁸

Table 2 Concentration of formaldehyde (ppb) in indoor air determined by the liquid droplet technique and reference method⁸

House		Cupboard
Liquid droplet technique	Reference method	Liquid droplet technique	Reference method
8.5	9.6	100	103
4.9	4.8	125	127
11.8	15.0	69.5	72.8
21.0	24.4	85.5	87.3

---

Validation of formaldehyde collection from air. Two methods for absorbing formaldehyde from air were adopted. The method with boric acid lacks good linearity and has poor precision. The RSDs for the formaldehyde absorbed from air by boric acid varied from 6 to 21%. The regression relation and the correlation coefficient between the concentration of formaldehyde in the wind tunnel and the concentration of formaldehyde in droplet were A = 10.5X − 211 (r² = 0.797). Therefore, in order to improve the collection efficiency of the trapping of formaldehyde from air by the droplet method, boric acid was mixed with hydralazine were for droplet correction. When boric acid and hydralazine used together, the efficiency of the method for formaldehyde was enhanced and the precision and accuracy were improved. The RSD varied from 3 to 12%, with a 2 μg ml⁻¹ detection limit. The regression equation and the correlation coefficient of linearity between the concentration of formaldehyde in the wind tunnel and the concentration of formaldehyde in droplet were A = 11.8X −53.1 (r² = 0.9283). The correlation curve is shown in Fig. 5. The collection efficiency of formaldehyde by the liquid droplet techniques is >90%.

Fig. 5 Correlation between the concentration of formaldehyde in the wind tunnel and in the liquid droplet.

Conclusion

The liquid droplet method provides a reproducible and accurate collector for gaseous analytes. The characteristics of the droplet may eliminate interferences from suspected particles without a special design. The droplet system is recyclable, renewable, consumes very little reagent and proceeds in a proper fashion. The main advantage of using a droplet system is that the measurement of the concentration of the compound in the droplet and in the atmosphere with appropriate techniques seems to be much simpler and easier. The droplet gradually increases in size when pumped at a flow rate of 0.010 ml min⁻¹. The formaldehyde gas was collected in a bottle and analysed by the proposed method.

The present method is sensitive, selective, simple and consumes a small amount of reagent, and the reagent itself is of low toxicity and its properties are well known, which is seldom true for reagents. Most of the interfering ions and species did not interfere in the determination. The excellent precision, freedom from pH control and elimination of an extraction process are major features of the developed method. Moreover, the limit of detection is also a significant advantage over other methods. In this study we have demonstrated the utility of the liquid droplet method for determining formaldehyde gases in indoor samples. The droplet device can be re-used. The droplet precision and accuracy were enhanced by using mixed boric acid and hydralazine.

Acknowledgements

The authors are grateful to the Japan Society for the Promotion of Science (JSPS) for a postdoctoral fellowship (P98448) awarded to Dr Helaleh and for financial support. This work was partly supported by Grants-in-Aid (10558085, 10650797) from the Ministry of Education, Science, Sports and Culture, Japan.

References

1. J. F. Walker, Formaldehyde, Reinhold, New York, 3rd edn., 1975..
2. M. P. Munoz, F. J. M. de Villena Ruenda and L. M. P. Diez, Analyst, 1989, 114, 1469.
3. S. M. Hays, R. V. Gobbell and N. R. Gauick, Indoor Air Quality: Solutions and Strategies, McGraw-Hill, New York, 1995, pp. 55–70..
4. Chemical Safety Data Sheets, Royal Society of Chemistry, Cambridge, 1990, Vol. 3, pp. 109–114..
5. J. J. Quackenboss, J. D. Spengler, M. S. Kanarek, R. Letz and C. P. Duffy, Environ. Sci. Technol., 1986, 20, 775.
6. E. Sawicki, T. W. Stanley and J. Pfaff, Anal. Chim. Acta, 1963, 28, 156.
7. A. P. Altshulter, D. L. Miller and S. F. Slega, Anal. Chem., 1991, 33, 621.
8. B. E. Hoogenboom, R. W. Hynes, C. M. Mann, M. Ekman, C. E. McJilton and J. B. Stevens, Am. Ind. Hyg. Assoc. J., 1987, 48, 420.
9. S. Velikonja, I. Jarc, L. Zupancic-Kralj and J. Marsel, J. Chromatogr., A, 1995, 704, 449.
10. J. D. Mulik, R. G. Lewis and W. A. McClenny, Anal. Chem., 1989, 61, 187.
11. R. Otson, P. Fellin, Q. Tran and R. Stoyanoff, Analyst, 1993, 118, 1253.
12. A. Kettrup, Analyses of Hazardous Substances in Air, VCH, New York, 1993, vol. 2, pp. 111–119..
13. A. Noda, K. Matsuyama, S.-H. Yen, N. Otsuji, S. Iguchi and H. Noda, Chem. Pharm. Bull., 1979, 27, 1939.
14. A. Noda, K. Matsuyama, S.-H. Yen, K. Sogabe, Y. Aso, S. Iguchi and H. Noda, Chem. Pharm. Bull., 1979, 27, 2820.
15. H. Noda, M. Minemoto, A. Noda, K. Matsuyama, S. Iguchi and T. Kohinata, Chem. Pharm. Bull., 1980, 28, 2542.
16. H. Noda, M. Minemoto, A. Noda and T. Ushio, Chem. Pharm. Bull., 1986, 34, 3499.
17. T. Tate, Philos. Mag., 1864, 27, 176.

---