[Home] [APPW 2004] [Journal papers]



Presented at the 59th Appita Conference, Auckland, New Zealand, 16-19 May 2005


Hexanal is an aldehyde that is supposed to originate as a reaction product from oxidation of linoleic acid which is a fatty acid component in wood extractives. Being one of the odour-contributing components in wood pulp, it is causing problems in, for example, food packaging board by bringing undesired off-flavour to the product The theory of the formation and removal of hexanal is discussed in this paper, as well as practical experiences. According to the results, certain conditions are key factors in formation as well as removal of hexanal.


Odour of paper may not have recognized industrial attention in the past. Today, it is a critical prerequisite for paper or board makers whose end product is intended for packaging where off-taint is unwanted. These are products intended to contain tobacco, chocolate, foodstuff etc. For example, a heavy-smoker can easily find if two cigarette packages are made during the same run/recipe simply by smelling the difference in the packaging.

There are several reasons for odour in paper: it can originate from anaerobic microbiological activity especially in connection with recycled fibres (1), the paper chemical itself or degradation products thereof (2), wood extractives and oxidation products thereof, ink etc. Discussion on these can be found in the literature, both theoretical and those based on trial experiences. However, the issue has not created a lot of interest in the past, but lately it has started to gain more attention due to everincreasing customer demands.

The odour of paper has been noticed to resemble that of fatty acid products, which has lead to the supposition that odour is emitted from auto-oxidation of fatty acids. A predominant aldehyde in pulp, hexanal, is one such general volatile organic compound that is proposed to be one indicator of paper odour (3). It is suggested to be an oxidation product of linoleic acid. In fact, there are other volatile aldehydes, ketones or alcohols formed and the reactivity of their precursors are different as well as the content in different kind of pulps (4). Volatile aldehydes ranging from C5 to C10 have low threshold values, giving taint to the product if the content of volatile compounds is above specifications.

The chemistry of auto-oxidation of linoleic acid is complex. To put it briefly, there is a generalized theory that oxidation of linoleic acid is an autocatalytic free radical reaction, possibly initiated by metals: unsaturated fatty acid produces a free radical by metals, which in turn produces a peroxyradical with oxygen and then the fatty acid hydroperoxide is degraded to produce hexanal. Hexanal is usually analyzed by headspace-GC, but the absolute content of it is heavily dependent on the analysis technique and sampling procedure. More detailed discussion on the analysis is found in the literature (3,5). Also sensory panel has been reported as a tool to assess odour (6,7,8), but its applicability may not always be practical compared to headspace-GC whereby one can better quantify the amount of volatiles. On the other hand, the result obtained by headspace-GC does not always translate to the human perception, but today it is well accepted to measure the content of hexanal by headspace- GC in order to assess the odour of pulps.

It has been discussed how to control the content of hexanal. Being volatile, it is supposed to simply disappear in the air in suitable conditions. On the other hand, in certain conditions, i.e. oxidative bleaching according to the proposed reaction mechanism, the content of hexanal in pulp should increase, because the oxidative chemicals will break the double bond in linoleic acid.

Several ways to reduce the hexanal content have been proposed. According to the reaction pathway, removal of metals from pulp by chelating agents should be beneficial because metals initiate the auto-oxidation. Also the use of antioxidants have been proposed or tested to have some effect (6,9). As an example, the use of antioxidants to inhibit auto-oxidation of lipids is well known in the food industry. Furthermore, extensive washing of the pulp as well as complete deresination, whether obtained by controlling the cooking or bleaching conditions, or by using extractive removal techniques like dispersants, should lower the content of fatty acids. Also removal of parenchyma cells in the screening room has been proposed as a way to reduce fatty acid content, which are enriched in these cells. In kraft pulping, SO2 is commonly used for acidification on drying machines or after the last bleaching stage, and it provides reducing conditions that can inhibit the oxidative reactions of linoleic acid to proceed.


The pulp used in this study was taken from a Finnish mill producing birch kraft pulp. Usually the content of extractives per ton pulp remains higher in mechanical pulps, but chemical pulps are a significant fibre source in packaging board, thereby having strong influence on the content of fatty acid in the pulp material used for board. However, certain chemi-mechanical pulps can exhibit very low hexanal content ranging from almost down to the level of chemical pulps (4). Also water recirculation in mills enhances build-up of extractives in a system. Nevertheless, kraft pulp remains the only object of pulps in this study. A wider experimental plan including mechanical pulps would have been too extensive for the scope of this study and comparison of hexanal content in different pulp types can be found elsewhere in the literature (4).

The study is based on the knowledge in the literature that hexanal derives from fatty acid but all the "how and why" of controlling it is not fully clear. In this context, we try to clarify how hexanal is formed in the kraft process, and how it can be removed. As hexanal is volatile and hence supposed to disappear gradually, the "competing reactions", formation and evaporation, are of interest. Thus for example storing of bleached pulp affects the content of hexanal in the final pulp. The removal is here focused on the way of utilizing reducing agents, mainly SO2. As an example, reductive dithionite bleaching of TMP has been shown to decrease the content of hexanal (10). Also, as SO2 is commonly used for acidification on drying machines or after the last bleaching stage, and provides reducing conditions that can inhibit the oxidative reactions of linoleic acid to proceed, the focus in this paper is to study how reducing agent can decrease the content of hexanal. Furthermore, hexanal is volatile and supposed to evaporate out from the pulp and hence also temperature is studied in this paper. More information on the experimental layout and the conditions are found in appendix 2.

In addition, formation of hexanal is supposed to be dependent on the oxidative conditions, therefore different oxidative bleaching chemicals were tested, namely peracetic acid, ClO2 and alkaline peroxide. The competing reaction was taken into account by using different sampling and storing techniques. Pulps were stored in three manners: a) open to light and air, b) in black bags to avoid light exposure and c) sealed in aluminium bags to avoid light and air. Pulps were stored for 4 weeks and the content of hexanal was measured during storage to see if there was an increase of hexanal content (suggesting oxidative conditions in pulp suspension due to residuals or alike), or a decrease (suggesting evaporation) or both. In addition, different treating times with reducing agent were tested as well as varying dosages. Detailed information on the experimental layout and the conditions are found in appendix 2.

The pulps were taken from an E1-stage decker, whereafter the pulp was washed and centrifuged and kept in freeze to avoid deterioration of the pulp. A ClO2 stage was made with the pulp under constant conditions, and then the final stage with ClO2, peracetic acid or peroxide under varying conditions. Pulps were bleached to full brightness 88%ISO. The conditions used in the study are highlighted in appendix 1.

Hexanal was measured according to the method presented in reference 2. The amount of pulp sample for the analysis of hexanal was 2g. Hexanal dissolved in triacetin was used as external standard and the content of hexanal was calculated from the calibration curve. A headspace-GC-MS was used with the settings presented in appendix 1.

All pulp bleaching and handling were made according to the acknowledged methods available. Other properties of the pulps were determined according to SCAN methods.


The effect of reducing agent and dosage

Figure 1 shows that there is actually no difference between acidifying or reducing agents, namely SO2 and NaHSO3 in the studied pH range. In fact, sulphur dioxide is in the form of bisulphite at pH 5,5 according to its phase diagram. Usually the drying machines run at pH 3,5-5,5. The small difference found in the third column is suggested to derive from slower reaction or improper mixing of bisulphite. It is also noted that drying at higher temperature seems to evaporate more of the hexanal away. An interesting phenomenon was found in the second column, as the content of hexanal increased after 3 weeks when drying in oven. This may be due to increased break-down of fatty acids to hexanal caused by higher temperature. These results help to undestand the phenomena happening in the end-section of a kraft pulp mill, namely from the last bleaching stage to pulp baling. This area is of special interest with regard to hexanal as there is high temperature zones (drying section), possibly residual chemicals present from the last bleaching stage, and more or less open light atmosphere.

Odour fig 1

Figure 1. Comparison of hexanal content in kraft pulp with SO2 and NaHSO3 under different drying conditions and storing time.


To clarify how much of the reducing agent is needed to lower the content of hexanal, figure 2 shows that 1kg SO2/tp could reduce hexanal significantly. Adding extra reducing agent did not improve the result. Probably this is due not to kinetics, but simply due to the reason that there are not more places available to react. During storing, the content of hexanal increased in open atmosphere (exposed to light and air) and the difference to SO2 treated pulps increased because of new formation of hexanal, indicating the formation is dependent on availability of light (discussed more later in this paper).

Odour fig 2

Figure 2. Effect of SO2 dosage on the content of hexanal. Pulp exposed to air and light during storing.


The effect of treatment time and storing conditions

Figures 3 and 4 compare the effect of time needed for SO2 treatment. It can be seen that increasing treatment time from 10 to 100 minutes improved the result, but already 10 minutes treatment gave a significant response. We tried also to go down to a quick 1 minute treatment, but there was too much scatter caused by incomplete mixing due to too short time (mixing in plastic bags by hand).

Odour fig 3

Figure 3. Effect of storing conditions on the content of hexanal, when pulp is treated with and without SO2 in 10 minutes.

Odour fig 4

Figure 4. Effect of storing conditions on the content of hexanal, when pulp is treated with and without SO2 in 100 minutes.


Also, when looking at the storing conditions, several conclusions could be drawn. When no SO2 was used and the pulps are stored securely in aluminium foil where no light or air is present, first hexanal increased in the beginning indicating formation of new hexanal and then the content started to decrease due to evaporation. This was similar with the pulps stored in bags that were exposed to air but not to light. Only in the case when the pulp was exposed to light and air, the content of hexanal proceeded to increase, which indicated that light is needed to initiate the auto-oxidation. If SO2 treatment was done, the content of hexanal remained lower all the way during storing, albeit only in the case of open storing conditions, new hexanal started to form again gradually later.

The effect of different bleaching chemicals on the storing behaviour

In figures 5 and 6, the effect of what bleaching chemical is used before SO2 treatment can be compared to figure 4. It can be concluded that the storing phenomena of hexanal, consecutively, is similar regardless if the prior bleaching step is conducted by ClO2, peracetic acid or peroxide. Again, the effect of sulphur dioxide on the removal of hexanal was seen in all series except the ones that were stored in the presence of air and light. The only difference between these bleaching chemicals was seen in the starting level of hexanal after the bleaching step prior to SO2 treatment. The higher starting level in the case of peracetic acid is due to the experimental layout: it is known that peracetic acid is among the most effective oxidative chemicals that can also easily break down extractives – here it was the most powerful oxidant, which is seen higher starting level of hexanal. The lowest starting level in the case of peroxide is most probably caused by the fact that alkaline peroxide stage dissolves and washes out extractives – there will simply be lower base level with the current experimental layout.

Odour fig 5

Figure 5. Effect of storing conditions on the content of hexanal, when pulp is bleached with peroxide before SO2 treatment.

Odour fig 6

Figure 6. Effect of storing conditions on the content of hexanal, when pulp is bleached with chlorine dioxide before SO2 treatment.


Other aspects and discussion

According to the theory presented in the introduction, metals should play a significant role in the formation of hexanal. Hence the metal content in the pulp should affect the hexanal formation. Also it is known that metals induce yellowing of pulp. Thus, we measured hexanal content and PC-number (yellowing) of the pulps when extra metals were added, in order to try to elucidate the formation and removal phenomena. The results are not presented though, because of inconsistency in the measurements and in fact no clear differences could be seen. One would at least expect that higher metal content would lead to higher hexanal formation and also yellowing.

One has to bear in mind though that these results apply to the experimental layout used here. On the other hand, the chemistry is the same always, but differences in the content of hexanal are most probably seen with different pulp and wood types, and with different processing techniques. Basically, one could expect that there is always both linoleic acid and hexanal present in a pulp, because the removal of extractives from pulp is never complete in practice and the same probably applies for evaporation of the formed hexanal. Ideally, to have the lowest content of hexanal in pulp, one should take use of extensive deresination techniques combined with oxidative bleaching and finally a reducing agent to maintain the level of hexanal as low as possible.


The content of hexanal in pulps was found to decrease with usage of sulphur dioxide. A reasonable reduction of almost half of the original amount was obtained in more than 10 minutes handling with SO2 at the addition of 1kg SO2 per ton pulp. During 4 weeks storing of pulp, the content of hexanal was found to increase if the pulp was stored in an open light and air atmosphere. This did not happen when the pulp was stored in the absence of light, indicating that the on-going formation of hexanal in pulps during storing is induced by light. Oxidative treatment was found to increase the content of hexanal.


The authors thank the laboratory personnel in Kemira's bleaching laboratory for their valuable contribution and Mrs. Mari Pispa for her skillful work on headspace-GCMS.


1. Dyer, J. – Progress in Paper Recycling 5(4):87-90(1996)

2. Lindell H. – Dissertation, Åbo Academy Press, Turku, Finland (1991)

3. Frisell, H. – Nordic Pulp and Paper Res. J. 17(1):74-78(2002)

4. Oyaas, K. - Nordic Pulp and Paper Res. J. 17(2):111-118(2002)

5. Forsgren, G., Norborg, M., Lindell, H., Räsänen, T., Frisell, H. – Nordic Pulp and Paper Res. J. 7(1):79-83(2002)

6. Wiik, K., Gromsrud, M., Helle, T. – Journal of Pulp and Paper Sc. (1998)

7. Rodbotten, M., Wiik, K., Helle, T. – 7-10 TAPPI Proceedings, Process & Product Quality Conference

& Trade Fair, (1997)

8. Wiik, K., Rodbotten, M., Bakken, T., Helle, T. – 11-17 TAPPI Proceedings, Process & Product Quality

Conference & Trade Fair, (1997)

9. Fagerlund, A., Shanks, D., Sunnerheim, K., Engman, L, Frisell, H. – Nordic Pulp and Paper Res. J.


10. Oyaas, K. – Tappi J. 2(7):25-28(2003)

Authors' contact

HANNU HÄMÄLÄINEN, Kemira Oyj, Vaasa, Finland

KALLE EKMAN, StoraEnso Oyj, Imatra, Finland

PETRI LASSILA, StoraEnso Oyj, Uimaharju, Finland

Odour appendix 1