This is the old United Nations University website. Visit the new site at http://unu.edu

Stability of iodine in iodized salt used for correction of iodine-deficiency disorders. II

Abstract
Introduction
Materials and methods
Results and discussion
Conclusions
References

L. L. Diosady, J. O. Alberti, M. G. Venkatesh Mannar, and S. FitzGerald

L. L. Diosady and J. O. Alberti are affiliated with the Department of Chemical Engineering and Applied Chemistry in the University of Toronto in Toronto, Ontario, Canada. M. G. Venkatesh Mannar is with the Micronutrient Initiative at the International Development Research Center in Ottawa, Ontario, and S. FitzGerald is affiliated with PATH (Programme for Appropriate Technology in Health) Canada in Ottawa.

Mention of the names of firms and commercial products does not imply endorsement by the United Nations University.

Abstract

The purpose of this study was to assess the effect of humidity and packaging materials on the stability of iodine in typical salt samples from countries with tropical and subtropical climates, under controlled climatic conditions. Initially we examined eight samples. In the second phase we expanded the study to salts from 18 sources and attempted to correlate the observed stability with salt impurities naturally present in these samples. High humidity resulted in rapid loss of iodine from salt iodized with potassium iodate, ranging from 30% to 98% of the original iodine content. Solid low-density polyethylene packaging protected the iodine to a great extent. High losses were observed from woven high-density polyethylene bags, which are often the packaging material of choice in tropical countries. Impurities that provided moisture at the salt surface had the most deleterious effect. Although clear correlations were not obtained, the presence of reducing agents, hygroscopic compounds of magnesium, and so forth seemed to have the most adverse effects on the stability of iodine. Surprisingly, carbonates had little effect on stability over the range present in the samples. Packaging salt in low-density polyethylene bags, which provided a good moisture barrier, significantly reduced iodine losses, and in most cases the iodine content remained relatively stable for six months to a year. The findings from this study indicate that iodine can be highly unstable, and in order to ensure the effectiveness of local salt-iodization programmes, countries should determine iodine losses from local iodized salt under local conditions of production, climate, packaging, and storage.

Introduction

Perhaps the greatest recent success of public health programmes has been the rapid expansion of salt-iodization programmes throughout the world during the past decade. The food supply of more than 1.6 billion people is lacking in adequate levels of iodine, resulting in the widespread prevalence of a spectrum of iodine-deficiency disorders, and the regular delivery of small doses of iodine to large populations through salt is beginning to have a dramatic effect. More than 50% of the world’s population now has access to iodized salt, and the virtual elimination of iodine-deficiency disorders in the world by the year 2000 has been set as a goal at several international forums, including the World Health Assembly [1], the World Summit for Children [2], and the International Conference on Nutrition [3].

Salt is an excellent carrier for iodine, as it is consumed at relatively constant, well-definable levels by all people within a society, independently of socio-economic status. Salt is iodized by the addition of fixed amounts of potassium iodide or iodate, as either a dry solid or an aqueous solution, at the point of production or import.

The actual availability of iodine from iodized salt at the consumer level can vary over a wide range as a result of:

» variability in the amount of iodine added during the iodization process;

» uneven distribution of iodine in the iodized salt, within batches and individual bags;

» losses of iodine due to salt impurities, packaging, and environmental conditions during storage and distribution;

» losses of iodine due to food processing, washing, and cooking processes in the household.

Iodization level

The stability of iodine in salt and levels of iodization are questions of importance to national planners and salt producers, as they have implications for programme effectiveness, safety, and cost. Higher levels of iodine may need to be added to compensate for losses due to known high levels of impurities in salt or the use of lower-grade packaging. This added cost must be compared with the cost of producing more stable, purified salt and the cost of enhanced packaging, while keeping in mind the consumer’s need for continuity in sensory qualities of the salt. Significant changes from the traditional products may result in higher costs to the producer and consumer, or reduced consumer acceptability, thus reducing the sustainability of the iodization programme.

Typical iodization levels vary from approximately 30 to 100 mg of iodine per gram of salt. In this work the salt was iodized with potassium iodate at a level of 50 mg of iodine per gram of salt - a value typical of many iodization programmes in tropical and subtropical countries.

Stability of iodine in salt

Elemental iodine readily sublimes and is then rapidly lost to the atmosphere through diffusion. Potassium iodide can be oxidized to elemental iodine by oxygen or other oxidizing agents, especially in the presence of catalysts, such as metal ions, and moisture. Thus, in affluent markets, iodide is always added to salt together with a reducing agent, such as dextrose, and a desiccant or anti-caking agent is usually included.

Potassium iodide can be reduced to elemental iodine by a variety of reducing agents in salt. Moisture naturally present in salt or abstracted from the air by hygroscopic impurities such as magnesium chloride acts as the reaction medium for the decomposition of added iodate. The pH of the condensed moisture on the salt is influenced by the type and quantity of impurities present, and this may in turn affect the stability of the iodine compounds. As in most chemical reactions, elevated temperature increases the rates of the reactions that form elemental iodine and increases the rate of evaporation of iodine.

Salt is extracted from a variety of sources, and the degree of purity depends on the source, extraction, and purification methods used. As a result, salt that is available for iodization may contain not only sodium chloride but also carbonate and sulphates, insoluble matter, and moisture. Physically, salt may be sold as large, crude crystals or as a refined, pure, dry powder.

On the basis of the chemistry, losses of iodine were not unexpected, and there have been a number of published and unpublished studies on the stability of iodine in salt. The relevant published literature was reviewed in the first phase of this programme [4]. The experiments conducted in the first phase clearly indicated that high humidity reduces stability, while the use of a good vapour barrier, which prevents the penetration of moisture and the evaporation of iodine, clearly improved the stability of iodine in iodized salt samples.

In this second phase of the study, the number of samples tested was expanded from the initial 8 to 18, including a Canadian salt as a reference sample. A sample from China was tested at two different levels of added iodine, as its iodine stability in the initial tests was unexpectedly low. The trace components of 23 samples were analysed, and attempts were made to correlate them with the observed iodine stability.

Packaging materials

Salt is sold in developing countries both in consumer packages of up to 2 kg and in bulk. Packaging materials in wide use in developing countries include paper, high- and low-density polyethylene, and woven bags made of jute, straw, or high-density polyethylene. The first phase of the study indicated that solid, non-woven polymer bags were the best moisture barriers and, if properly sealed and intact, would maintain the moisture level of the salt throughout the distribution system, thus minimizing the loss of iodine following the absorption of moisture and subsequent chemical reactions. Accordingly, a solid film of low-density polyethylene and a woven bag of high-density polyethylene were used in the second phase of the study.

Objectives

The purpose of the second phase of the study was to assess the stability of iodine in typical salts, in an effort to identify the causes and extent of iodine loss. The effects of humidity, purity of the salt, and packaging materials on the stability of iodine in typical salt samples from countries with tropical and subtropical climates were determined under controlled climatic conditions. In the short term, the results may be useful for assessing potential losses of iodine from salt between the points of production and consumption. The overall goal of the study was to determine the range and timing of iodine losses that may be expected under typical conditions when iodine is added to salt in the form of iodate, and thus to define the most cost-effective means of controlling or compensating for these losses, to ensure that populations at risk for iodine-deficiency disorders receive effective amounts of iodine from iodized salt.

Materials and methods

Materials

Potassium iodate, analytical reagent grade, was obtained from BDH, Toronto.

Samples of un-iodized salt, such as that consumed by low-income populations, were obtained from 12 countries: Bangladesh, Bolivia, China (two levels of addition), Ghana, Guatemala, India (five sources), Indonesia (four sources), Pakistan, Philippines (three sources), Senegal, Thailand, and Tanzania. The samples were obtained through UNICEF country offices from routine production runs from local salt producers and air shipped to Toronto.

A sample of un-iodized refined salt obtained from Toronto Salt Chemical Co. was also tested. This Canadian sample was used as a reference.

Sample treatment

Salt samples with particle sizes less than 2 mm were used without pre-grinding. Because wet salt could not be sieved, salt samples containing large particles and having a moisture content over 3% were dried in a forced-convection oven at 70°C overnight, ground by mortar and pestle, and sieved to pass through a 10-mesh sieve. The water content was then reconstituted to the original moisture level.

Two-kilogram samples of salt from each source were fortified to contain about 50 mg/kg iodine using potassium iodate added as a 30 g/L solution. The mixtures were blended to ensure uniformity using a 5-L ribbon blender (LeRoy Somer, LS Tronics, Montreal).

Packaging materials

In the second phase of the programme, two packaging methods were tested, as the use of open containers was shown to be impractical in the first phase of the programme. For each treatment condition, two 500-g samples were prepared in solid (continuous-film), low-density polyethylene bags, 0.07 mm in thickness, and in woven, high-density polyethylene bags, 0.15 mm in thickness. The bags were folded closed but were not sealed. Both low-density polyethylene and high-density polyethylene are clear, transparent or translucent, plastic materials that are extruded as a thin sheet. High-density polyethylene has a much higher tensile strength, as it is made of longer molecular chains. Low-density polyethylene was made into bags by folding the sheets into appropriate shapes and welding the seams by heating, whereas high-density polyethylene bags were made by cutting the sheets into thin strips 1.5 to 2 mm wide and weaving them into a “cloth,” which was then sewn to form bags. Even though high-density polyethylene does not absorb water, the woven bags readily allow the passage of water through the weave.

Storage conditions

The packages were stored under two conditions: elevated temperature and high humidity (40°C, 100% relative humidity) and elevated temperature and medium humidity (40°C, 60% relative humidity).

High-temperature, high-humidity conditions were maintained by using a controlled-temperature oven in which the air was saturated with water vapour by exposure to a tray of water. High-temperature, medium-humidity conditions were maintained in an environmental chamber manufactured by Associated Environmental Systems Division of Craig Systems Corporation (Toronto). Temperature and humidity were automatically monitored.

Analytical methods

Sampling

Packages of salt were sampled at the start of the experimental series (denoted as 0 months in the results) and after 1,2, 3, 6, and 12 months of storage. To obtain a representative and reasonably homogeneous sample for analysis, the complete solid salt content of a bag was split into two equal subsamples by pouring it through a two-stemmed powder funnel. The splitting of the sample was repeated until only about 15 g of salt was collected. This subsample was used for the analyses.

Moisture

The moisture content was determined gravimetrically. Samples of salt were weighed and then dried at 110°C for 16 hours and reweighed.

Iodine

Iodine content was measured by neutron activation analysis under titration.

Neutron activation analysis

Neutron activation analysis is a non-destructive testing method in which a sample is irradiated at high ray flux in a nuclear reactor, and the specific radiation from selected radioisotopes formed by the irradiation is measured. This technique has a high relative standard deviation (~5%), but it is not subject to interference from reducing or oxidizing agents [5]. Therefore it was used to confirm the results of colourimetric measurements in each sample series.

1. Approximately 1.25 g of salt was accurately weighed into a polyethylene vial. To decrease interference due to the presence of a high concentration of chlorine in the sample, the sample was shielded with cadmium.

2. The vials were irradiated at 1 kW power with a neutron flux of 5.0 X 10¹¹ cm^-2 s^-1 for three minutes using the University of Toronto’s SLOWPOKE nuclear reactor.

3. The samples were removed from the reactor and rested for six minutes.

4. After six minutes of delay, gamma emission at 443 keV was measured using a hyperpure germanium-based gamma ray spectrometer.

5. The iodine content was calculated using a calibration curve established by a series of spiked samples that covered the range from 5 to 250 mg of iodine per kilogram of salt.

The titrimetric analytical method for iodate is more rapid and convenient and significantly less expensive than the neutron activation analysis method. However, there is a potential for large positive or negative errors if the salt contains significant amounts of oxidizing or reducing impurities. All salts were first analysed by neutron activation, but subsequent analyses were done by titration if there was no significant difference between the iodine values obtained by the two techniques.

1. Ten grams of salt was dissolved in approximately 100 ml of water. The pH was adjusted to 2.8 using 0.6% HCl.

2. Thirty milligrams of potassium iodide powder was added (12-fold excess) to convert all of the iodate present to elemental iodine by the reaction:

KIO₃ + 5KI + 3H₂O -> 3I₂ + 6KOH

3. The liberated iodine was titrated with 0.004 N freshly prepared sodium thiosulphate solution. Starch was used as the end-point indicator. The relative standard deviation of the analysis was determined to be <1%.

Salt impurities

The typical impurities present in salt were determined in each of the salt samples when they were received. The sampling and sample-preparation methods described above were used. The range of analyses is described below.

Elemental analyses

Calcium, magnesium, and iron were measured by atomic absorption spectrophotometry in the absorption mode, using a Perkin-Elmer Model 704 instrument (Perkin-Elmer Canada, Toronto, Ontario). All of the other reported elements were analysed by an inductively coupled argon plasma spectrometer at the facilities of Zenon Environmental (Burlington, Ontario).

Carbonates

Carbonates and bicarbonates were determined by the AOAC standard titrimetric method 33.071 [6].

Results and discussion

All salt samples lost iodine over the 12-month sampling period. Losses ranged from less than 10% to 100% of the original iodine in the sample (from a starting value of 50 mg of iodine per gram of salt + 5%). The rate of iodate loss was influenced by the salt’s origin and composition, the packaging material, and the relative humidity during storage.

Relative humidity

The study was not designed to reproduce the storage conditions of any specific locality. The study’s objective was to test extreme conditions that may be encountered in typical distribution systems. As expected, storage of salt samples at 100% relative humidity resulted in the greatest iodine loss. Even under moist, tropical conditions, it is unlikely that the relative humidity would remain at this extreme level for a long period. However, within bags exposed to sunlight, or in storage facilities heated by the sun, the high humidity will be retained, once moisture is absorbed into the bag’s contents, and temperatures may readily rise to over 60°C. Only one month of exposure to 40°C and 100% relative humidity resulted in most of the samples in high-density polyethylene bags losing more than 25% of the added iodine.

In all cases, the samples stored at 60% relative humidity lost iodine at a lower rate than those stored in saturated air (100% relative humidity). After six months, storage losses at 60% relative humidity ranged from ~0% to 20%, which is lower than might be expected [7], and losses after 12 months averaged approximately 40%.

At high humidity the losses were more dramatic. Iodine losses over six months of storage ranged up to 100%, indicating that within the 12-month trial period, essentially all of the iodine added to the sample disappeared from high-density polyethylene bags, which are effectively open to the atmosphere.

Packaging material

In the first stage of the study, we demonstrated that packaging type had a significant effect on storage stability. In the second phase of the work, only low-density polyethylene film and high-density polyethylene woven bags were used. Solid low-density polyethylene provided an excellent moisture barrier and thus maintained the total water content of the bags approximately constant, near the level at the time of packaging. Some absorption of moisture was possible, as the bags were not sealed sufficiently to prevent some diffusion of air containing water and/or iodine into and out of the bags.

Woven high-density polyethylene bags readily allowed the moist air to contact the salt and thus release iodine as a vapour, and they also allowed any condensed moisture to drip out of the bag in the form of a saturated salt solution containing iodate.

The effect of relative humidity and type of container on the average values of iodine retention in all 18 samples tested is presented graphically in figure 1.

Salt purity

The physical characteristics of salt samples, which were affected by the profile of impurities and the extent of processing at the source, had a major effect on the stability of the salt. The salt samples received varied in colour from very bright white to dark gray or rusty brown. The samples had particle sizes ranging from -100 mm to 30 mm, with great variability in the homogeneity of particle size. A description of all samples is presented in table 1.

A wide range of trace components was found in these samples (table 2). The sample from Guatemala must have been iodized, as it contained 22 mg of iodine per gram of salt. The titrimetric method revealed that it was in the form of iodate.

Two other samples from Indonesia and Pakistan contained oxidizing agents equivalent to 7 and 8 ppm iodate iodine, although the samples contained < 1 mg of iodine per gram of salt by neutron activation analysis. Most of the salt samples contained trace amounts of iodine (1-2 ppm by neutron activation analysis). However, this was not a surprise, as earlier studies found similar trace levels in purified sea salt from the United States. This low concentration of iodine cannot be detected by the titrimetric method.

The salt sample from China lost all of its iodine very rapidly, even though it was of high sensory quality, and contained few impurities in significant quantities. Indeed, even at time = 0 a significant loss from the added iodate was observed by the titration method. In an effort to check and quantify this unexpected result, we repeated the test with the addition of 100 mg of iodine per gram of salt as potassium iodate. This sample also showed a rapid initial loss of iodine, but then the iodine level stabilized.

Titration of the un-iodized Chinese salt showed that it contained a reducing agent capable of immediately freeing 14 mg of iodine per gram of salt. On standing, this value would increase slowly, thus accounting for the complete loss of iodine from the addition level of 50 mg of iodine per gram of salt. Subsequently we found that the processor counteracted this effect by adding iodine as potassium iodide. Potassium iodide is actually stabilized by the impurity, which is probably a sulphur compound.

FIG. 1. Effect of relative humidity and container type on iodine retention. Abbreviations: LDPE, low-density polyethylene film bag; HDPE, woven high-density polyethylene bag; RH, relative humidity

Canadian salt, which was of high purity and contained very little moisture or hygroscopic impurities, was relatively stable. At low humidity the iodine losses were less than 10% during the first six months of storage and less than 25% after a year. At 100% humidity the protection of the polyethylene bag maintained iodine losses at less than 8%, whereas in woven high-density polyethylene bags, 63% of the iodine was lost.

Some salt samples with lower levels of purity and higher levels of moisture were as stable as the Canadian sample. Whereas the salts lost, on average, 33% of their iodine after 12 months at 60% relative humidity in low-density polyethylene, salts from Canada, India, the Philippines, and Senegal lost less than 15% of their iodine after 6 months, even at 100% relative humidity in low-density polyethylene. Only the samples from China lost more than one third of their iodine after six months, yet the average iodine retention dropped to 44% of the amount added at the end of one year of storage. The Tanzanian salt retained only 22% of its iodine after a year under these conditions.

TABLE 1. Physical characteristics of salt samples

Sample no.	Source	Appearance of salt	% insoluble matter	Appearance of insoluble matter
1	Tanzania	Very hard crystals, some brownish-black impurity	2.5	Brownish, claylike substance
2	Bolivia	Very large (3-4 cm) pinkish crystals	0,7	Brown, sandy residue
3	Indonesia (P.T. Garam)	White crystals, some “dirt” sandy residue	1.6	Black, claylike substance with brown,
4	China (Beijing)	White, fine crystalline powder < 1 mm	1.2	Fine, brown-black dust
5	Ghana	White crystals 2-5 mm	1.8	Black and brown, sandy material
6	India (Arumuganeri)	Fine white powder <1 mm	2.0	Fine to coarse, brown, sandy material
7	Philippines (code PHI-MI4)	Brownish crystals 2-4 mm	2.0	Black and yellow clay
8	Senegal	White crystals 2-8 mm	0.9	Yellow, sandlike material
9	Canada	White, fine, crystalline powder <1 mm	0.4	No visible residue
10	China repeat	White, fine, crystalline powder <1 mm	1.2	Brown and black dust particles
11	India (Kurkuch)	Brownish, sandlike, 2-4 mm crystals	2.8	Fine to coarse, dark brown, sandy material
12	Pakistan	Slightly brownish crystalline solids	2.5	Light brown, fine dust with white particles ~2 mm)
13	India (Phoda)	Very large, pinkish crystals 5-15 mm	0.7	Brown, claylike material
14	Guatemala	Brown, sandlike powder 2-4 mm	1.7	Black, clay-type residue
15	Indonesia (Central Java)	Yellow crystals 3-8 mm	2.0	Brown, black, clay-type substance
16	Philippines (code PHI-PCS)	White, clean crystals 2-4 mm	0.7	Brown, clay-type material
17	Thailand	White, crystalline powder <1 mm	1.9	Brownish-yellow, clay-type material
18	Bangladesh	White crystals 2 - 6 mm	0.7	Brown, clay-type and yellow sandy particles
19	India (Tuticorin)	White, crystalline powder < 1 mm	2.3	Brownish-yellowish sand- and soil-like residue
20	India (“crushed” sign)	White to brown crystals 2-4 mm	1.9	Black, brown clay-type residue
21	Indonesia (Madura)	Yellow crystals 5-15 mm	3.3	Brown, black clay-type substance
22	Indonesia (West Java)	Yellowish-brown crystals 3-8 mm	0.7	Brown, clay-type substance
23	Philippines (code PHI-B2M)	White, crystalline powder <1 mm	1.7	Fine to coarse, dark, sandy material

TABLE 2. Chemical characteristics of salt samples

Sample no.	Source	pH	Moisture (%)	Carbonates (ppm)	Bicarbonates (ppm)	Total carbonates (ppm)	Calcium (ppm)	Magnesium (ppm)	Barium (ppm)	Potassium (ppm)	Iron (ppm)	Strontium (ppm)	Sulphur (ppm)
1	Tanzania	9.77	11.5	375	488	863	1,400	4,600	6.9	1,800	<50	48	3,500
2	Bolivia	8.48	2.1	10	443	453	2,600	380	<0.2	<500	<50	77	2,400
3	Indonesia (P.T. Garam)	8.78	6.0	35	442	477	1,100	3,300	<0.2	1,300	<50	57	2,500
4	China (Beijing)	8.58	0.2	23	361	384	910	260	<0.2	<500	<50	20	840
5	Ghana	8.32	9.5	5	430	435	2,500	1,700	<0.2	<500	<50	62	2,800
6	India (Arumuganeri)	8.28	2.1	2	454	456	3,800	540	2.3	<500	<50	260	3,200
7	Philippines (code PHI-MI4)	9.35	6.4	245	513	758	3,500	4,100	<0.2	<500	<50	71	4,700
8	Senegal	8.87	3.6	65	432	497	1,900	3,500	<0.2	1,400	<50	84	3,200
9	Canada	6.25	0.4	0	41	41	290	210	<0.2	<500	87	3	160
10	China repeat	8.58	0.2	23	361	384	910	240	<0.2	<500	<50	20	840
11	India (Kurkuch)	7.54	0.7	0	318	318	3,100	700	<0.2	<500	<50	37	2,700
12	Pakistan	9.08	6.2	83	1,396	1,479	5,800	4,300	<0.2	<500	75	180	12,000
13	India (Phoda)	7.86	0.8	0	287	287	2,600	690	<0.2	<500	<50	30	2,200
14	Guatemala	8.72	4.5	50	452	502	6,300	4,000	<0.2	1,600	<50	140	6,700
15	Indonesia (Central Java)	7.88	3.9	0	413	413	2,200	1,900	<0.2	<500	62	48	2,600
16	Philippines (code PHI-PCS)	9.07	3.2	118	671	789	3,100	4,000	<0.2	1,400	<50	85	4,600
17	Thailand	7.94	1.4	0	468	468	750	1,900	<0.2	<500	<50	45	1,600
18	Bangladesh	7.21	1.5	0	825	825	1,100	1,100	<0.2	<500	<50	32	1,300
19	India (Tuticorin)	8.62	3.6	19	363	382	3,000	3,000	<0.2	<500	<50	120	2,400
20	India (“crushed” sign)	7.48	1.8	0	379	379	3,700	740	<0.2	<500	<50	47	3,200
21	Indonesia (Madura)	7.78	5.0	0	1,256	1,256	1,700	3,400	<0.2	1,100	<50	66	3,000
22	Indonesia (West Java)	8.14	2.7	102	470	572	910	2,900	<0.2	<500	<50	42	2,200
23	Philippines (code PHI-B2M)	8.79	13.0	118	671	789	3,300	9,900	<0.2	2,700	<50	110	7,600

The stability of the samples was significantly lower in high-density polyethylene bags. Whereas at low humidity the losses were, on average, reasonable after six months (18%), after a year the loss rose to 37%. At high humidity the losses were much more significant. Seven of the samples lost half of their iodine content after only one month, and the average iodine retention decreased rapidly to 37%, 10%, and 1% after 3,6, and 12 months, respectively. The Indian Kurkuch and the Pakistani samples retained more than 90% of the original iodine at three months but only 61% and 28%, respectively, after six months. None of the samples retained more than 12% of the iodine after one year. The results are presented in tables 3 to 6.

There was no clear and consistent correlation between iodine stability and the presence of any impurity. Clearly there are many competing reactions and interactions between the salt impurities and the added potassium iodate. There is a trend towards lower iodine retention with increased magnesium and sulphur content, but the sparing effect of carbonates was not observed at the levels present in the samples. The stability dropped markedly in the first three months when samples were stored at high humidity and the magnesium content exceeded 1,000 ppm. After six months at high temperature and humidity, all samples lost most of their iodine independently of magnesium content, whereas at low humidity all samples were relatively stable (fig. 2).

The effect of pH was also not clear-cut. The data indicated that the stability was lower around pH 8.0, but this effect was not statistically significant. The effects of magnesium and pH are shown in figures 2 and 3.

Conclusions

The study clearly indicates that moisture plays a critical role in the stability of iodine. In particular, when salt is stored at a temperature characteristic of storage and distribution conditions in many developing countries, moisture absorbed by hygroscopic impurities contributes to the rapid loss of iodine. Although the loss of iodine cannot be clearly correlated with any specific impurity on the basis of the present study, it is clear that the use of highly purified salt would improve the stability of iodine in most cases. Unfortunately, this would not be technically or economically feasible in the short term in many developing countries.

TABLE 3. Iodine retention in low-density polyethylene film bags at 40°C and 60% relative humidity

Sample	Origin	% of original iodine remaining after storage for:
		1 mo	3 mo	6 mo	12 mo
1	Tanzania	95.1	97.8	83.9	70.3
2	Bolivia	89.2	95.3	85.3	68.7
3	Indonesia (P.T. Garam)	98.9	93.7	96.5	79.0
4	China (Beijing)	13.5	3.2	0	0
5	Ghana	93.7	98.9	100.0	88.9
6	India (Arumuganeri)	100.0	98.7	99.1	87.4
7	Philippines (code PHI-MI4)	100.0	100.0	100.0	82.5
8	Senegal	93.8	93.2	93.8	76.1
9	Canada	95.6	94.2	94.9	77.3
10	China repeat (100 ppm)	57.1	54.3	44.5	39.5
11	India (Kurkuch)	100.0	100.0	91.5	69.6
12	Pakistan	89.2	79.4	76.4	69.6
13	India (Phoda)	100.0	87.3	76.1	69.8
14	Guatemala	100.0	88.6	79.3	74.9
15	Indonesia (Central Java)	85.6	71.4	65.6	60.9
16	Philippines (code PHI-PCS)	98.0	95.8	92.0	67.9
17	Thailand	100.0	100.0	91.2	63.6
18	Bangladesh	93.0	82.6	77.9	67.8
	Average	89.0	85.2	80.4	67.4

TABLE 4. Iodine retention in woven high-density polyethylene bags at 40°C and 60% relative humidity

Sample	Origin	% of original iodine remaining after storage for:
		1 mo	3 mo	6 mo	12 mo
1	Tanzania	89.4	91.5	89.2	69.3
2	Bolivia	92.1	95.0	98.4	68.7
3	Indonesia (P.T. Garam)	95.9	94.3	98.2	78.5
4	China (Beijing)	5.4	8.3	0	0
5	Ghana	92.2	90.8	99.6	81.0
6	India (Arumuganeri)	100.0	99.1	98.0	75.7
7	Philippines (code PHI-MI4)	98.6	98.4	99.6	78.8
8	Senegal	91.8	90.5	90.3	77.1
9	Canada	100.0	94.9	94.9	76.4
10	China repeat (100 ppm)	55.3	51.2	41.1	37.2
11	India (Kurkuch)	100.0	100.0	95.8	67.5
12	Pakistan	86.7	81.0	74.0	67.4
13	India (Phoda)	98.2	93.6	88.9	71.8
14	Guatemala	100.0	97.5	85.2	59.1
15	Indonesia (Central Java)	80.9	68.9	62.0	37.0
16	Philippines (code PHI-PCS)	96.6	93.4	93.0	66.5
17	Thailand	94.9	88.7	87.4	58.5
18	Bangladesh	100.0	91.3	77.9	63.5
	Average	87.7	84.9	81.9	63.0

By packaging salt in an effective moisture barrier, such as solid low-density polyethylene bags, iodine losses can be significantly reduced. With solid low-density polyethylene packaging, the losses of iodine from salt stored for up to six months can be kept in the range of 10% to 15%, but the losses generally increase significantly over the next six months of storage, and therefore the time required for distribution, sale, and consumption should be minimized to ensure effective use of the added iodine.

To retain the storage efficiency of low-density polyethylene films in a system of high mechanical strength and resistance to puncture, woven high-density polyethylene bags with a continuous film insert or laminate of low-density polyethylene should be investigated as an effective, low-cost packaging method for iodized salt.

The results indicate that the control of moisture content in iodized salt throughout manufacturing and distribution by improved processing, packaging, and storage is critical to the stability of the added iodine. In order to make allowances for the probable losses of iodine, countries must determine iodine losses from local iodized salt under local conditions, as these will be greatly affected by the quality of the packaged salt.

TABLE 5. Iodine retention in low-density polyethylene film bags at 40°C and 100% relative humidity

Sample	Origin	% of original iodine remaining after storage for:
		1 mo	3 mo	6 mo	12 mo
1	Tanzania	88.0	89.6	72.2	22.4
2	Bolivia	91.7	95.0	67.8	41.4
3	Indonesia (P.T. Garam)	93.2	91.2	77.5	46.2
4	China (Beijing)	14.7	9.9	12.5	0
5	Ghana	95.9	100.0	72.8	17.9
6	India (Arumuganeri)	100.0	100.0	91.5	56.8
7	Philippines (code PHI-MI4)	99.3	99.3	89.0	62.4
8	Senegal	92.5	93.4	87.6	63.8
9	Canada	100.0	96.0	92.0	66.6
10	China repeat (100 ppm)	56.0	54.8	55.4	45.6
11	India (Kurkuch)	99.6	91.9	73.8	44.0
12	Pakistan	84.3	77.0	64.7	59.0
13	India (Phoda)	95.8	78.1	70.0	52.3
14	Guatemala	97.9	82.7	65.0	33.7
15	Indonesia (Central Java)	75.7	71.9	66.3	33.2
16	Philippines (code PHI-PCS)	94.2	90.6	74.9	55.0
17	Thailand	88.9	88.0	69.0	37.1
18	Bangladesh	100.0	86.1	78.1	50.7
	Average	87.1	83.1	71.1	43.8

TABLE 6. Iodine retention in woven high-density polyethylene bags at 40°C and 100% relative humidity

Sample	Origin	% of original iodine remaining after storage for:
		1 mo	3 mo	6 mo	12 mo
1	Tanzania	51.0	12.0	2.4	0
2	Bolivia	85.8	64.9	9.2	0
3	Indonesia (P.T. Garam)	91.3	27.1	7.2	0
4	China (Beijing)	26.6	14.5	7.5	0
5	Ghana	92.2	90.9	61.4	3.9
6	India (Arumuganeri)	98.7	93.3	28.1	1.1
7	Philippines (code PHI-MI4)	92.0	36.7	3.4	0
8	Senegal	91.2	14.1	1.6	0
9	Canada	89.7	72.2	4.2	2.0
10	China repeat (100 ppm)	48.8	40.1	16.0	12.2
11	India (Kurkuch)	87.9	29.0	8.7	0
12	Pakistan	37.0	6.5	4.0	0
13	India (Phoda)	20.1	5.4	3.0	0
14	Guatemala	92.2	58.0	12.3	0
15	Indonesia (Central Java)	13.4	12.7	7.1	0
16	Philippines (code PHI-PCS)	39.4	6.2	0	0
17	Thailand	78.6	71.7	7.1	0
18	Bangladesh	81.0	7.4	3.5	0
	Average	67.6	36.8	10.4	1.1

FIG. 2. Effect of magnesium on iodine retention in salt from India stored at 40°C. Abbreviations: LDPE, low-density polyethylene film bag; HDPE, woven high-density polyethylene bag; RH, relative humidity

FIG. 3. Effect of pH on iodine retention in salt from India stored at 40°C. Abbreviations: LDPE, low-density polyethylene film bag; HDPE, woven high-density polyethylene bag; RH, relative humidity

Acknowledgements

This project was a collaborative effort between PATH Canada, the Department of Chemical Engineering and Applied Chemistry of the University of Toronto, and the Micronutrient Initiative. We gratefully acknowledge the financial assistance of the Micronutrient Initiative and the assistance of UNICEF and UNICEF field offices in procuring representative salt samples.

References

1. Word Health Organization. Handbook of resolutions and decisions of the World Health Assembly. 39th World Health Assembly (WHA39-31, 1986) and 43rd World Health Assembly (WHA43.2,1990). Geneva: WHO, 1993.

2. UNICEF. First call for children and plan of action from the world summit for children/convention on the rights of the child. New York: UNICEF, 1990.

3. US Agency for International Development. World declaration and plan of action for nutrition. International Conference on Nutrition. Rome: Food and Agriculture Organization/World Health Organization, 1992.

4. Diosady LL, Alberti JO, Venkatesh Mannar MG, Stone TG. Stability of iodine in iodized salt used for correction of iodine-deficiency disorders. Food Nutr Bull 1997;18:388-96.

5. Kruger P. Principles of activation analysis. New York: Wiley Interscience, 1971.

6. Association of Official Analytical Chemists. Official methods of analysis. 14th ed. Methods 33.147 and 33.175. Washington, DC: AOAC, 1984.

7. Mannar MGV, Dunn JT. Salt iodization for the elimination of iodine deficiency. The Netherlands: ICCIDD, 1995.