Evaluation of the potential effects of ingredients added to cigarettes. Part 2: Chemical composition of mainstream smoke

Abstract

Cigarette mainstream smoke from blended research cigarettes with and without the addition of ingredients was analyzed for its chemical composition. In total, 333 ingredients commonly used in cigarette manufacturing were assigned to three different groups. Each group of ingredients was introduced at a low and a high level to the test cigarettes. The list of the 51 smoke constituents determined is based on those analytes suggested for analysis in a US Consumer Product Safety Commission proposal for low ignition cigarettes and cigarette smoke constituents identified by the International Agency for Research on Cancer as worthy of concern and characterized as carcinogens. An increase in the yield of total particulate matter (TPM) in the range of 13 to 28% relative to the control cigarette without ingredients was observed for all test cigarettes. This was presumably caused by the higher transfer rates of the added ingredients to the smoke compared to the transfer from the tobacco part of the filler. When the yields of individual constituents were normalized to the TPM yields, a reduction in the majority of the constituents was observed when compared to the control. For one of the ingredient groups this reduction was especially high: for phenols a maximum of 70%, for polycyclic aromatic hydrocarbons 50%, and for N-nitrosamines 45%. An increase in the amount relative to TPM was observed for a few smoke constituents: hydrogen cyanide and cadmium (one ingredient group), formaldehyde (one ingredient group), and resorcinol and lead (two ingredient groups). These results are consistent with the lack of any increased activity in the in vitro and in vivo assays in this same series of studies (Food and Chemical Toxicology 2002, 40, 105–111; Food and Chemical Toxicology 2002, 40, 113–131). An overall assessment of our data suggests that these ingredients, when added to the tobacco, do not add to the toxicity of smoke, even at the elevated levels tested in this series of studies.

Introduction

Ingredients added to tobacco can undergo different fates during cigarette smoking. Depending on their physical and chemical properties, various amounts of a compound may be distilled, combusted, or pyrolysed into different products. As most of the ingredients used are highly volatile aroma compounds, they are transferred unchanged into mainstream smoke (Green, Chalmers, & Kinnard, 1989, Jenkins, Newman, & Chavis, 1970). Typical examples of these ingredients are menthol, anisole, benzyl alcohol, vanillin and certain essential oils. Many of these compounds or mixtures are also natural constituents of the tobacco leaf. Because most ingredients are added to the tobacco in very small amounts (a few parts per million) and because of the low transfer into mainstream smoke, their contribution to mainstream smoke composition is consequently also very low. Other typical ingredients are compounds of low volatility such as sugars, and fruit juice extracts, which contain predominantly sugars and aroma compounds. These low volatile ingredients are expected to behave essentially like the sugars naturally contained in the tobacco leaf and to contribute in a similar way to mainstream smoke composition (Gager, Nedlock, & Martin, 1971, Gager, Nedlock, & Martin, 1971).

The fate of some of the individual ingredients investigated in this study has been analysed previously, either by combusting/pyrolysing the pure ingredient or in a burning cigarette. Examples of the first case are investigations into the pyrolysis of cocoa (Schlotzhauer et al., 1985) and the pyrolysis of carbohydrates (Tomasik et al., 1989). Examples of the second case are the investigations of volatile flavor compounds (Green, Chalmers, & Kinnard, 1989, Jenkins, Newman, & Chavis, 1970) and humectants (Laurene et al., 1965). The concept of the present study varies from the previous ones in that all ingredients were tested by adding them in groups to a single research cigarette. The design covers possible interactions between ingredients and/or their combustion/pyrolysis products by combining as many ingredients as possible in a single cigarette type. No attempt was made to determine the total recovery and identification of individual ingredients or their specific combustion/pyrolysis products. Instead, this study investigated the contribution of these processes to the overall delivery of mainstream smoke constituents.

The cigarettes investigated in this study have already been described in detail in Part 1 of this publication series (Carmines, 2002). The test cigarettes each contain one of three different groups of ingredients in a low or high level, and the control cigarette contains no added ingredients. The low level approximates the normal use level considered to be reflective of those used in commercial cigarettes and the high level is either 1.5 or 3 times the low level. The test cigarettes were designed to burn in a comparable manner, and this limitation prohibited the use of extremely high levels of the ingredients.

In this study, approximately 50 smoke constituents, selected predominantly for their toxicological properties (Voncken et al., 1998), were determined with validated analytical procedures to study the influence of the addition of ingredients on the composition of mainstream smoke.