Skip to main content
  • Research article
  • Open access
  • Published:

Analysis of hydrazine in smokeless tobacco products by gas chromatography–mass spectrometry



Due to the lower health risks associated with the use of certain categories of smokeless tobacco products (STPs) such as Swedish snus, there is interest in the comparative levels of toxic chemical constituents in different types of STPs. A method has been developed and validated for the analysis of hydrazine in STPs. Seventy four commercial STPs from the US and Sweden, representing 80-90% of the 2010 market share for all the major STP categories in these two countries, as well as three reference STPs, were analysed for hydrazine.


Aqueous extracts of the STPs were treated with excess pentafluorobenzaldehyde (PFB), which reacted with hydrazine in solution to form decafluorobenzaldehyde azine (DFBA). DFBA was partitioned into hexane and then quantified by gas chromatography–mass spectrometry (GC–MS). The method was validated using five different types of STP, was linear in the range 8–170 ng/mL, and had limits of quantification (LOQ) from 26–53 ng of hydrazine per g of STP (as sold). The method was applied to the analysis of 74 contemporary STPs commercially available in the United States and Sweden, none of which were found to contain hydrazine above the LOQ or LOD. Trace levels of compounds showing chromatographic and mass spectral features consistent with hydrazine were identified at very low levels (sub-limit of detection, <10 ng/g) in the chromatograms of less than half of the 74 STPs examined; in contrast, for 40 of the STPs no evidence for the presence of hydrazine was observed. Where present, the levels of compounds consistent with hydrazine were estimated to be at least an order of magnitude lower than the only previous study to have quantified hydrazine in tobacco.


Our results show that hydrazine is not a prevalent constituent of STPs, and when present is not quantifiable using currently available analytical methodology.


Smokeless Tobacco Products (STPs) have been collectively designated as Group 1 carcinogens i.e. carcinogenic to humans [1], but there is considerable evidence that health risks differ between STP categories, with certain product styles such as Swedish snus having lower health risks associated with their use [2]. As a result, there is substantial interest in the comparative levels of toxic chemical constituents of the different types of STPs. In the US, the Food and Drug Administration (FDA) has assembled a list (“The Established List”) of 93 “hazardous or potentially hazardous constituents” (HPHC) of tobacco products which may have to be reported [3]. This list covers both tobacco and tobacco smoke components and includes 79 that are designated as carcinogenic as well as constituents that are respiratory toxicants, cardiovascular toxicants, reproductive toxicants or addictive. One of the HPHC carcinogens on the list is hydrazine (N2H4) which has been classified as a group 2B carcinogen (possibly carcinogenic to humans) by IARC [4].

Although studies of hydrazine toxicity in humans are limited, human exposure to hydrazine has resulted in severe effects on the central nervous system, liver and kidneys [4]. Hydrazine is mainly an industrial chemical, manufactured from ammonia, that can enter the environment from facilities that manufacture, process or use it. Hydrazine is unstable and degrades rapidly in most environmental media. It can dissolve in water and move though soil, but hydrazine is broken down by autoxidation and by microorganisms. A review [5] of three studies concluded that half-lives of hydrazine in soil ranged from 1 hr to 3 days with the more rapid degradation of hydrazine occurring in soils with high levels of microorganisms and organic material.

The occurrence of free hydrazines in nature is rare. Naturally occurring hydrazine and hydrazone derivatives such as agaritine (β-N-[γ-L(+)-glutamyl]-4-hydroxymethyl phenylhydrazine) and gyromitrin (acetaldehyde methylformylhydrazone) have been reported in mushrooms [6]. Gyromitrin breaks down during cooking to release methylhydrazine but the latter is not found in a free state in the mushroom. Hydrazine is produced as an intermediate during biological nitrogen fixation by the molybdenum- and vanadium-based nitrogenase enzymes in Azotobacter [7]. In the case of the more abundant molybdenum-based nitrogenase the hydrazine is bound to the enzyme and is not released in a free state. However for vanadium- based nitrogenase small but significant amounts of free hydrazine are generated [8]. There is therefore the potential for hydrazine to be found in plant materials that are associated with nitrogenase containing bacteria.

The only occurrence of free hydrazine in plant material was that reported in tobacco by Liu et al. [9]. They found small amounts of hydrazine in tobacco from a commercial cigarette (30.0 ng/cigarette) and in the tobacco of four experimental cigarettes. Two of the experimental cigarettes were made with Burley tobaccos, one treated with the plant sucker growth inhibitor maleic hydrazide (MH) (51.2 ng hydrazine/cigarette) and one untreated (22.2 ng hydrazine/cigarette); the other two cigarettes were made with flue-cured tobacco, one treated (12.1 ng hydrazine/cigarette) and one untreated (13.8 ng hydrazine/cigarette). Liu et al. [9] also determined hydrazine in the mainstream smoke of these 5 cigarettes (range 23.5–42.8 ng/cigarette). The hydrazine concentrations in tobacco and tobacco smoke obtained in the original Liu et al. study of over 40 years ago [9], have been frequently reproduced in review articles [10-16]. No other study of hydrazine in tobacco has been reported, although several other studies have failed to detect hydrazine in tobacco smoke [17-19]. Using the same methodology as Liu et al. for hydrazine, Schmeltz et al. [13] found the hydrazine derivative, 1,1-dimethylhydrazine, in several samples of tobacco including US chewing tobacco (97.7 ng/g) and snuff (96.7 ng/g), four commercial US cigarette blends (60.2 ± 5.7 ng/g) and Bright tobacco (147 ng/g). No 1,1-dimethylhydrazine was found in a sample of Burley tobacco.

We are currently conducting a comprehensive survey of toxicants in an extensive and varied set of contemporary STPs from the United States and Sweden. There have been no further published studies of hydrazine in tobacco since the report by Liu et al. in 1974, and no studies of hydrazine in STPs have ever been reported. The aims of the present study were therefore to develop and validate a method for the analysis of hydrazine in STPs, and to survey major STPs from the USA and Sweden for their hydrazine content.

Various methods have been reported for the detection of trace levels of hydrazine in substrates such as sludge, human plasma, environmental water and drug samples. These have included chemical derivatization with reagents such as benzaldehyde, 4-hydroxy benzaldehyde, 2-hydroxynaphthaldehyde, 2,4-dinitrochlorobenzene and acetone or acetone-d6, coupled with HPLC and/or spectrophotometric detection [20-23], solid-phase spectrophotometry [24], and GC–MS [25]. Indirect methods have also been developed, such as oxidation of hydrazine by excess iodate [26] or bromine [27], and analysis of the unreacted oxidant. The detection limits for these methods vary from 1 ng/ml [21] to 100 ng/g [25]. The analytical method used in the original study in tobacco by Liu et al. [9] was based on the reaction of residues of hydrazine in tobacco product extract with PFB to form DFBA, Figure 1, coupled with multiple thin-layer chromatographic steps, followed by GC separation and electron capture detection.

Figure 1
figure 1

Reaction of pentafluorobenzaldehyde (PFB) with hydrazine to form decafluorobenzaldehyde azine (DFBA).

For the analysis of STPs, we chose to use derivatization of hydrazine with PFB, as described in the Liu et al. paper, but coupled with GC–MS to improve the sensitivity and to eliminate the need for the multiple concentration steps used in the original paper. The present method was based on the reaction of residues of hydrazine in tobacco product extract with PFB to form DFBA. The DFBA is partitioned into hexane and then quantified by GC–MS. After validation of this approach, the method was applied to the analysis of 74 contemporary STPs commercially available in the United States and Sweden. The products covered all major STP categories and the brands selected represented 90% market share of the major product styles [28].

Results and discussion

Validation of the analytical method

As a first step in validating the analytical approach, the identity of the peak assigned to DFBA was confirmed by visual examination of the chromatograms and mass spectra obtained for standards and tobacco samples spiked with hydrazine at 0.53 μg/g. The retention time of the GC peak assigned to DFBA was 9.9 min for all standards and types of STP. The mass spectra of the assigned peaks were almost identical for all standards and spiked tobacco samples and included ion clusters at m/z 388 (molecular ion and base peak), 194, 180, 117 and 93. Lastly, the MS software selected the spectrum of DFBA from the “Saturn library” as the closest match to that of the chromatogram peak.

To check the linearity of the method, six standard DFBA solutions, ranging from 100 to 2000 ng/mL (equivalent to 8–170 ng/mL hydrazine), were applied to GC–MS in a random order. The areas of the peaks were subjected to linear regression analysis. The analysis was highly linear across the six standards with R 2 values in excess of 0.99 (Figure 2).

Figure 2
figure 2

Linearity of the analytical method. The intercept on the x axis represents a hydrazine concentration of 0.0018 μg/mL.

In a preliminary analysis, the five control tobacco samples (Garrett, Stonewall Wintergreen, Ettan Loose, Days Work and Taylor’s Pride), each representing a different type of STP, were analysed in duplicate. All of these STPs had apparent hydrazine contents below the lowest standard analysed (100 ng/mL DFBA), corresponding to concentrations of <16 ng/g.

To test the accuracy and precision of the analytical method, analyses were carried out in which the five control STPs were spiked with hydrazine at 530 ng/g, 53 ng/g and 26.5 ng/g. For each STP at each level of spiking, five repeat analyses were carried out (Table 1).

Table 1 Recovery and repeatability of measurements of spiked tobacco samples

At spike levels of 530 and 53 ng/g hydrazine, the mean recoveries from all tobacco product types were within the acceptable range [29] of 70%–110% (Table 1), indicating satisfactory accuracy. At the spike level of 26.5 ng/g hydrazine, a low recovery of 64% was observed for Stonewall Wintergreen. The hard pellet has a high mineral content that may have affected the hydrazine added, either through chemical reaction or adsorption. With the %RSD below 20%, the precision of the analytical technique was satisfactory at all three spike levels.

The LOQ was defined as the lowest spike level for which there was an acceptable recovery (i.e. in the range 70%-110%). The LOQ was therefore 26.5 ng/g for the chewing tobacco, dry snuff, moist snuff and loose snus, and 53 ng/g for the hard pellet. The limit of detection (LOD) was estimated to be less than 10 ng/g from this validation. The linearity, LOQ, LOD, precision, and accuracy of the method are summarized in Table 2.

Table 2 Linearity and sensitivity of the analytical method for hydrazine

Survey of 74 STPs for hydrazine

Once validated, the method was used to survey the 74 contemporary STPs for hydrazine levels. As shown in Tables 3 and 4, none of the products was found to contain hydrazine at levels above the LOD (<10 ng/g). Examination of the chromatograms identified a very low level peak at the retention time of hydrazine (Figure 3A, C), and with a matching mass spectrum, for 34 of the 74 STPs analysed. The peaks were sufficiently infrequent, and below the LOD, that the possibility cannot be discounted that these peaks were merely analytical noise. However, as the peaks showed identical chromatographic and mass spectral features to hydrazine, we regard it as possible that very low levels of hydrazine were present in the samples showing these peaks. Of these STPs, the majority (22) showed the hydrazine peak in only one of the three replicates analysed, seven STPs showed the peak in two of the three replicates, and five STPs showed the peak in all three replicates. In total 51 of the 222 replicate analyses showed the presence of hydrazine, and therefore the majority of analyses showed no evidence for the presence of low levels of hydrazine (Figure 3B, D, E). Clearly, if there is interest in quantifying these potential very low levels of hydrazine, a much more sensitive analytical method would be required, with more than an order of magnitude greater sensitivity. It is unlikely that the current approach is modifiable to this extent, and alternative approaches may be required.

Table 3 Estimated hydrazine concentrations in Swedish STPs
Table 4 Estimated hydrazine concentrations in US STPs
Figure 3
figure 3

Typical chromatograms for smokeless tobacco products in the analysis of hydrazine (retention time 10.0 minutes); A) Camel Frost Snuff, B) Camel Mellow Snuff C) Marlboro Mild Snuff, D) Marlboro Peppermint Snuff, E) Marlboro Rich Snuff.

As noted above, in every case that hydrazine was tentatively identified in the current study the peak areas were substantially below the LOQ and LOD of the current method, and therefore the levels present cannot be determined. However, overall, our results indicate that hydrazine is not a prevalent contaminant of contemporary STPs, and in the minority of cases where a peak consistent with hydrazine was observed, the levels present are substantially lower than those reported previously by Liu et al. [9].

Sources of hydrazine in tobacco

Liu et al. [9] considered the possibility that the MH used as a sucker growth inhibitor on the tobacco crop was the source of hydrazine observed in their tobacco. Hydrazine is a contaminant in MH that derives partly from the manufacturing process and partly from subsequent breakdown of MH (particularly the formulation conjugated with diethanolamine, MH-30) [30]. However Liu et al. [9], albeit on a limited number of samples, found no relationship between MH concentrations and hydrazine. MH treated samples of tobacco had similar levels of hydrazine to samples containing no MH. Subsequent to the Liu et al. study, the diethanolamine salt was banned (in 1980) and only the more stable potassium salt of MH is currently approved for use. The US EPA [31] and the European Union [32] have also introduced limits on the concentration of hydrazine in MH - 15 ppm in the US and 1 ppm (1 μg/g) in the EU. Using the CORESTA issued Guidance Residue Levels (GRL) on agrochemicals of 80 ppm for MH on tobacco [33] as an upper limit, and assuming no hydrazine losses from the tobacco post MH-application, it can be calculated that maximum concentrations of hydrazine in tobacco arising from contaminated MH would be 1.2 ng/g in the US and 0.08 ng/g in the EU. The current study therefore does not rule out the possibility that breakdown of MH might contribute to traces of hydrazine in the tobacco.

An alternative to MH as a source of hydrazine in tobacco was advanced by Schmeltz et al. [13]. They reported the hydrazine derivative, 1,1-dimethylhydrazine, in several samples of US tobaccos using the same methodology as Liu et al. [9]. Schmeltz et al. [13] proposed that unspecified bacterial and enzymatic processes that occur during curing might be responsible for producing both the 1,1-dimethylhydrazine observed in their study and also the hydrazine observed in the earlier Liu et al. study. To date, however, there have been no reports of microorganisms or enzymatic pathways, such as nitrogenase, specifically related to tobacco that would result in hydrazone or hydrazine formation.

Differences between levels reported by Liu and results of the present study

Whatever the source of the hydrazine it is unlikely that it would be present in tobacco in a free state. Hydrazine is a powerful reducing agent, and reacts with carbonyls to form azines and hydrazones as shown in Figure 4 [34]. STPs and other forms of tobacco have been shown to contain ppm levels of formaldehyde, acetaldehyde, crotonaldehyde [10,35,36], acrolein [36], acetone, propionaldehyde, isobutyraldehyde, 2-butanone isovaleraldehyde and valeraldehyde [35]. Although the presence of carbonyl azines or hydrazones in tobacco has not been reported [37], given the thousand-fold excess of carbonyls over hydrazine in tobacco, there is a strong possibility that any hydrazine present in the tobacco plant will react with the carbonyls present. The observation that hydrazine reacts rapidly with acetaldehyde in tobacco smoke [38] adds to the plausibility that the same reaction can occur in the tobacco matrix. Moreover hydrazine is relatively volatile with a boiling point of 114°C and it would seem probable that any unreacted hydrazine would be lost during processing particularly for STPs such as snus where there is sustained heat treatment of the tobacco.

Figure 4
figure 4

Reaction of ketones (R 1 , R 2 = alkyl) and aldehydes (R 1 = alkyl, R 2 = H) with hydrazine to form azines (A) and hydrazones (B).

Liu et al. [9] pointed out that the highly reactive complexing agent PFB not only reacts with any free hydrazine in the matrix but is also able to react with any hydrazones or azines that may be present. They demonstrated this by showing that more than 70% of the hydrazine moiety of benzalazine was detected as pentafluorobenzaldehyde azine (PFBA) during controlled experiments over a 16 hour period representative of their experimental extraction conditions for tobacco and smoke (Figure 5). Thus we would expect that any hydrazones and azines present in the tobacco matrix could also react with PFB. Hence the hydrazine content of tobacco reported by Liu et al. may well have been the sum of hydrazine, hydrazones and azines present.

Figure 5
figure 5

Reaction of benzaldehyde azine (benzalazine) with pentafluorobenzaldehyde (PFB) to form pentafluorobenzaldehyde azine (PFBA) and decafluorobenzaldehyde azine (DFBA).

In the present study a much shorter contact time between tobacco and the reactive complexing agent PFB was used. The one hour complexation time used in the present study is an order of magnitude shorter than the “overnight” time used by Liu et al. [9]. Notably, the hydrazine contents identified in this work are an order of magnitude lower than reported by Liu et al. [9]. One explanation for the difference in contents could therefore be a restricted opportunity for reaction of PFB with hydrazones or azines in this work compared to that in the study of Liu et al. [9].


Tobacco samples

Tobacco samples were obtained in 2010. Details of the STP markets in the United States and Sweden were obtained, and the products for analysis were chosen to reflect approximately 90% market share of the major STP categories in these two markets at that time. The major products in each category of STP were sampled. In total, the survey comprised 31 Swedish products (10 loose snus and 21 portion snus, Table 5) and 43 US products (13 chewing tobaccos, 5 dry snuffs, 2 hard pellet products, 1 soft pellet product, 15 moist snuffs, 6 US snus and 1 plug product, Table 6). The Swedish products were sourced from Swedish retail websites, imported into the United Kingdom, and kept frozen at −20°C until analysis. The US products were sourced from shops in the United States, imported, and kept frozen at −20°C until analysis. Three CORESTA reference STP products [39] were also sampled and analysed in this exercise, CRP1 (Swedish snus pouch product), CRP2 (US style loose moist snuff) and CRP3 (US style loose dry snuff powder).

Table 5 Swedish STPs
Table 6 United States STPs


Hexane (SpS grade) and technical grade methanol were obtained from Romil Ltd (Cambridge, UK). 1.0 N hydrochloric acid, 2,3,4,5,6-pentafluorobenzaldehyde (PFB) (purity 98%), anhydrous sodium sulphate (≥99.0%), acetic acid (purity ≥99.0%), and hydrazine sulphate (purity 99%) were obtained from Sigma-Aldrich (Gillingham, Dorset, UK). The calibration standard, decafluorobenzaldehyde azine (DFBA) (purity 99.8%), was prepared following the method of Liu et al. [9]. The purity of DFBA was characterised by GC/MS and Differential Scanning Calorimetry.

GC–MS conditions

GC–MS analyses were performed using a Varian 3800–Saturn 4D GC–ion trap mass spectrometer, coupled with a Varian CP-8400 autosampler and a Saturn GC–MS Workstation running Star software version 5.51 and the following conditions: column, 30 m × 0.35 mm × 0.25 µm Zebron ZB-5 capillary column; oven program, 70°C increased to 250°C at 15°C/min, followed by a 3-min dwell time (15-min run time); injection temperature, 200°C; transfer line temperature, 220°C; manifold temperature, 250°C; injection volume, 2 µl; injection, splitless; helium flow rate, 1.0 mL/min; acquisition rate, m/z 40–550; filament delay, 5 min; quantitative ions, m/z 388 + 369 (corresponding to the molecular ion C14F10N2+ and a loss of F). An example chromatogram, for a standard solution of DFBA, is presented in Figure 6.

Figure 6
figure 6

Typical GC–MS chromatogram for a 0.5 μg/ml decafluorobenzaldehyde azine (DFBA) standard (equivalent to 0.04 μg/ml hydrazine). Chromatogram shows signal from mass 388.

Preparation of standards and tobacco samples for method validation

The DFBA standard was characterized by GC–MS and the purity was determined by differential scanning calorimetry. A stock standard solution, nominally 1000 µg/mL, of DFBA in hexane, was prepared in accordance with Liu et al. [9], and diluted with hexane to give standard solutions in the range 0.20 to 2.0 µg/mL of DFBA. The solutions were tested and found to be stable at room temperature for 4 weeks.

A stock solution of hydrazine sulphate, nominally 500 µg/mL of hydrazine in deionised water, was prepared for the recovery experiments. The exact concentration of hydrazine in the stock solution was calculated using the formula:

$$ \frac{Weight\ of\ hydrazine\ sulphate\ (g)\times purity\ of\ hydrazine\ sulphate\ \left(\%\right)\times 32.05\times {10}^6\ \mu g.m{L}^{-1}}{10^4\times 130.12} $$

where 32.05/130.12 is the conversion factor for hydrazine sulphate to hydrazine. The solution was stable at room temperature for 4 weeks. Diluted solutions of 1.0 to 10 µg/mL of hydrazine in deionized water were freshly prepared for recovery experiments.

Five different types of STP were used for the validation experiments: loose snus (Ettan, Swedish Match), dry snuff (Garrett, Conwood), chewing tobacco (Taylor’s Pride, Conwood), hard pellet (Stonewall Wintergreen, Star Scientific), and plug (Day’s Work, Swedish Match). The samples were extracted in hexane as follows. Two grams of STP were added to 50 mL of 20:80 methanol:0.1 N hydrochloric acid (aq.), and the mixture was incubated at room temperature for 1 hour. For finely divided tobacco and pellet samples, the flask was placed in an ultrasonic bath for 10 min; for leafy tobacco products the mixture was macerated with a blender for 10 min. After centrifugation for 5 min at 3000 rpm, 25 mL of supernatant was added to 0.2 mL of acetic acid and 10.0 mL of 1.0% PFB in methanol, and the mixture was swirled and allowed to stand at room temperature for 1 hour. Next, 20 mL of hexane was added, and the mixture shaken for 1 min. The aqueous phase was run into a clean flask, and the hexane fraction was passed over anhydrous sodium sulphate. The aqueous phase was extracted again by the above procedure with 20 mL of hexane, and then again with 10 mL of hexane, and the three extracts were combined. The combined hexane extract was reduced to approximately 1 mL using a rotary evaporator, transferred to a 2-mL volumetric flask, and made up to 2 mL with fresh hexane, ready for application to GC–MS.

The amount of hydrazine in each tobacco sample was calculated using the formula:

$$ \frac{Test\ sample\ area \times std.\ conc.\ \left(\mu g.m{L}^{-1}\right)\times 50\times 2\times 32.05\times 1000}{Mean\ std.\ area\times 2\times 25\times 388.17} $$

where 32.05/388.17 is the conversion factor for DFBA to hydrazine. Values in this study are reported in units of ng/g, as the common portion sizes of use are gram sized [40].

For recovery experiments, 2 g of STP was added to 100 µl of a standard hydrazine solution (0.5 μg/g, 0.05 μg/g and 0.025 μg/g hydrazine). The mixture was swirled and allowed to stand at room temperature for 10 min to enable the hydrazine to penetrate the matrix. Next, 50 mL of 20:80 methanol:0.1 N hydrochloric acid (aq.) was added and the tobacco sample extracted as described above. Accuracy was determined as the mean recovery ± relative standard deviation (RSD). Precision was determined as the repeatability RSD at each level.

Water content

STP samples were analysed for water content using Karl Fisher Coulometric analysis with an KEM MKC-500 analyser (Kyoto Electronics, Tokyo, Japan). Approximately 2 g STP was weighed into a 25 ml snap-top vial. 20.0 ml MeOH was added and the sample sonicated for 15 minutes before being allowed to steep and settle for at least 2 hours. 100 μl of methanol was sampled and injected into the Karl Fisher analysis cell. Water blanks were subtracted, and analyses conducted in triplicate.


In this study we have developed and validated a method to determine levels of hydrazine in tobacco, using PFB as a complexing agent to form the azine, DFBA, which was then quantified by GC–MS. In a series of 74 brands of smokeless tobacco from the US and Sweden covering all major product styles - snus, chewing tobacco, moist snuff, dry snuff, plug and pellet products - hydrazine concentrations were all below the level of quantification of 26.5 ng/g product. Peaks consistent with hydrazine were identified, at trace levels (below limit of detection), in the chromatograms of less than half of the 74 STPs, but at levels considerably lower than the only previous study to have quantified hydrazine in tobacco. All previous references to the presence of hydrazine in tobacco and STPs are based on a single study from 1974 [9] which measured the hydrazine contents of tobacco from five different cigarettes (4 experimental and 1 commercial).

There is insufficient information to explain the differences in results between the present study and the earlier study. However the much lower levels of hydrazine found in the present study are consistent with the reductions in maximum levels of hydrazine residues permitted in MH, which were legislated subsequent to the original study of Liu et al. Also our method used shorter times for complexation of hydrazine residues with PFB. The shorter complexation time may have minimised artifactual formation of DFBA through complexation with non-hydrazine moieties such as hydrazones and azines as Liu et al. reported as probable in their study. Nevertheless, the current study demonstrates that the presence of hydrazine in contemporary STPs is relatively infrequent, and when present hydrazine is at levels below the limits of quantification and detection that can be achieved using current analytical best-practice.



Below the limit of detection


Cooperation centre for scientific research relative to tobacco


Decafluorobenzaldehyde azine


US food and drug administration


Gas chromatography - mass spectrometry


Guidance residue level


Harmful and potentially harmful constituents


International agency for research in cancer


Limit of detection


Limit of quantification


Maleic hydrazide




Pentafluorobenzaldehyde azine


Relative standard deviation


Smokeless tobacco product


  1. IARC. Smokeless tobacco and some tobacco-specific N-nitrosamines. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, IARC Press, Lyon, France. 2007;89:370.

    Google Scholar 

  2. Royal College of Physicians. Harm reduction in nicotine addiction: helping people who can’t quit. London: RCP: A report by the Tobacco Advisory Group of the Royal College of Physicians; 2007.

    Google Scholar 

  3. FDA: Draft guidance for industry: reporting harmful and potentially harmful constituents in tobacco products and tobacco smoke under the Federal Food, Drug, and Cosmetic Act. 2012. []

  4. IARC. Re-evaluation of some organic chemicals, hydrazine and hydrogen peroxide. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, IARC Press, Lyon, France. 1999;71:991–1013.

    Google Scholar 

  5. Environment Canada/Health Canada. Screening Assessment for the Challenge. Hydrazine. Chemical Abstracts Service Registry Number 302-01-2. January 2011.

  6. Toth B. Synthetic and naturally occurring hydrazines as possible cancer causative agents. Cancer Res. 1975;35:3693–7.

    CAS  Google Scholar 

  7. Thorneley RNF, Eady RR, Lowe DJ. Biological nitrogen fixation by way of an enzyme-bound dinitrogen-hydride intermediate. Nature. 1978;272:557–8.

    Article  CAS  Google Scholar 

  8. Dilworth MJ, Eady RR. Hydrazine is a product of dinitrogen reduction by the vanadium-nitrogenase from Azotobacter chroococcum. Biochem J. 1991;277:465–8.

    CAS  Google Scholar 

  9. Liu YY, Schmeltz I, Hoffmann D. Chemical studies on tobacco smoke. XXVII. Quantitative analysis of hydrazine in tobacco and cigarette smoke. Anal Chem. 1974;46:885–9.

    Article  CAS  Google Scholar 

  10. Hoffmann D, Adams JD, Lisk D, Fisenne I, Brunnemann KD. Toxic and carcinogenic agents in dry and moist snuff. J Natl Cancer Inst. 1987;79:1281–6.

    CAS  Google Scholar 

  11. Brunnemann KD, Hoffmann D. Chemical composition of smokeless tobacco products. In Smoking and Tobacco Control Monograph. 1993;2:96–108.

    Google Scholar 

  12. Schmeltz I, Hoffmann D. Nitrogen-containing compounds in tobacco and tobacco smoke. Chem Rev. 1977;77:295–311.

    Article  CAS  Google Scholar 

  13. Schmeltz I, Abidi S, Hoffman D. Tumorigenic agents in unburned processed tobacco: N-nitrosodiethanolamine and 1,1-dimethylhydrazine. Cancer Lett. 1977;2:125–32.

    Article  CAS  Google Scholar 

  14. Hecht SS. Cigarette smoking and cancer. In: Rom WN, editor. Environmental and occupational medicine. New York: Lippincott-Raven; 1998. p. 1479–99.

    Google Scholar 

  15. Hatsukami DK, Ebbert JO, Feuer RM, Stepanov I, Hecht SS. Changing smokeless tobacco products. New tobacco-delivery systems. Am J Prev Med. 2007;33 Suppl 6:S368–78.

    Article  Google Scholar 

  16. Scientific Committee on Emerging and Newly Identified Health Risks. Health effects of smokeless tobacco products. 2008

  17. Patterson JM, Issidorides CH, Groutas VC, Smith Jr WT. Pyrolysis of maleic hydrazide and MN’dimethyldodecylamine. London: Chem Ind (London); 1972. p. 337–8.

    Google Scholar 

  18. Pailer M, Sulm O: Pyrolyse von Tabakadditiven. Fachliche Mitt Oesterr Tabakregie. 1973;14:258–266.

  19. Harke H-P, Schueller D, Drews C-J. Pyrolyse von maleinsӓurehydrazid und hydrazin. Z Lebensm Unters-Forsch. 1973;153:163–9.

    Article  CAS  Google Scholar 

  20. Elias G, Bauer WF. Hydrazine determination in sludge samples by high-performance liquid chromatography. J Sep Sci. 2006;29:460–4.

    Article  CAS  Google Scholar 

  21. Kirchherr H. Determination of hydrazine in human plasma by high-performance liquid chromatography. J Chromatogr. 1993;617:157–62.

    Article  CAS  Google Scholar 

  22. Khuhawar MY, Rajper AD, Rind FMA. High performance liquid chromatographic determination of hydrazine in water using 2-hydroxynaphthaldehyde as a derivatizing reagent. Pak J Sci Ind Res. 2002;45:371–4.

    CAS  Google Scholar 

  23. George M, Nagaraja KS, Balasubramanian N. Spectrophotometric determination of hydrazine. Talanta. 2008;75:27–31.

    Article  CAS  Google Scholar 

  24. Ortega-Barrales P, Molina-Díaz A, Pascual-Reguera MI, Capitán-Vallvey LF. Solid-phase spectrophotometric determination of trace amounts of hydrazine at sub-ng ml−1 level. Anal Chim Acta. 1997;353:115–22.

    Article  CAS  Google Scholar 

  25. Sun M, Bai L, Liu DQ. A generic approach for the determination of trace hydrazine in drug substances using in situ derivatization-headspace GC–MS. J Pharm Biomed Anal. 2009;49:529–33.

    Article  CAS  Google Scholar 

  26. Haji Shabani AM, Dadfarnia S, Dehghan K. Indirect spectrophotometric determination of trace quantities of hydrazine. Bull Korean Chem Soc. 2004;2004(25):213–5.

    Google Scholar 

  27. George M, Nagaraja KS, Balasubramanian N. Spectrophotometric determination of hydrazine using bromine and methyl red. Indian J Chem. 2007;46A:1621–4.

    CAS  Google Scholar 

  28. McAdam K, Faizi A, Kimpton H, Porter A, Rodu B. Polycyclic aromatic hydrocarbons in US and Swedish smokeless tobacco products. Chemistry Central. 2013;7:151.

    Article  Google Scholar 

  29. The Codex Alimentarius Volume 3. Residues of veterinary drugs in foods. 2nd ed. Rome Italy: Joint FAP/WHO Food Standards Programme, FAO; 1993. p. 59.

    Google Scholar 

  30. Bakker H, Martyn A, Schreuder RH. Gas–liquid chromatographic determination of hydrazine in maleic hydrazide formulations and in samples stored at an elevated temperature. Pestic Sci. 1983;14:470–4.

    Article  CAS  Google Scholar 

  31. US Environmental Protection Agency. Maleic hydrazide: Notification of issuances of notice of intent to suspend pesticide registration. Document OPP38501; PH-FRL-1934. Fed Regist. 1981;46(179):45999–6000.

    Google Scholar 

  32. European Commission: Health & Consumer Protection Directorate-General: Review report for the active substance maleic hydrazide. SANCO/10501/2002-final 2 December 2002.

  33. CORESTA, 2010,

  34. March J: Advanced Organic Chemistry 4th Edition, Hoboken, New Jersey: Wiley Interscience; 1992, p.904.

  35. Weybrew JA, Stephens RL. Survey of the carbonyl contents of tobacco. Tob Sci. 1962;6:53–7.

    CAS  Google Scholar 

  36. Stepanov I, Jensen J, Hatsukami D, Hecht SS. New and traditional smokeless tobacco: comparison of toxicant and carcinogen levels. Nicotine Tob Res. 2008;10:1773–82.

    Article  CAS  Google Scholar 

  37. Rodgman A, Perfetti TA. The chemical components of tobacco and tobacco Smoke. Second Edition: CRC Press; 2013.

    Book  Google Scholar 

  38. Harward CN, Parrish ME, Plunkett SE, Banyasz JL, Shafer KH. Evaluation of hydrazine reduction by cellulose acetate filters using infrared tunable diode laser spectroscopy. Anal Chem. 2002;74:5871–81.

    Article  CAS  Google Scholar 

  39. CORESTA, 2010,

  40. Digard H, Errington G, Richter A, McAdam K. Patterns and behaviors of snus consumption in Sweden. Nicotine Tob Res. 2009;11:1175–81.

    Article  Google Scholar 

Download references


We would like to acknowledge the contributions of GC Laboratories Ltd., Hitchin, UK to method development and sample analysis activities.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Kevin McAdam.

Additional information

Competing interests

The study was funded by British American Tobacco (BAT). KM, HK, CV, PD and CW are employees of BAT, as was SE at the time of the study. AP is a paid consultant to BAT. BR’s research is part funded by unrestricted grants from Tobacco Manufacturers (including BAT) to the University of Louisville.

Authors’ contributions

KM – co-directed the study and co-wrote the manuscript. HK – project managed the study. AP - co-wrote the manuscript. CV – identified, sourced and summarised hydrazine literature. SE, PD, CW – guided and managed the method development and validation. BR - co-directed the study and contributed to writing the manuscript.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits use, duplication, adaptation, distribution, and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

McAdam, K., Kimpton, H., Essen, S. et al. Analysis of hydrazine in smokeless tobacco products by gas chromatography–mass spectrometry. Chemistry Central Journal 9, 13 (2015).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: