Study of Colouring, Aromatic Strength and Bitterness of Saffron (Crucos sativus L.) Cultivated in the UK
A. Yadollahi, Z.A. Shojaei / A. FarahnakySchool of Biosciences
Sutton Bonington Campus
University of Nottingham
UK / Department of Food Science and TechnologySchool of Agriculture
ShirazUniversity
Shiraz
Iran
Keywords:APCI-MS, colourimetery, UV-visible spectroscopy
Abstract
Ideally, planting of saffron in new areas is a key resolving question of farmers which this article is a contribution of premier source of needed information about quality of saffron in the UK. Colouring and aromatic strength and bitterness of saffron (Crucos sativus L.) cultivated for the first time in theEast Midlands in the UK were determined by different methods and compared with those values of Iranian and Indian saffrons. Different approaches were performed including colourimetery (Hunter Lab), UV-visible spectroscopy (ISO 3632-2003) and APCI-MS (Atmospheric Pressure Chemical Ionisation- Mass Spectroscopy). The results showed that the colouring parameters (L, a, b and chroma values) of saffron powder were relatively lower than of those parameters of 1%w/w saffron in water for all samples. The chroma (an indication of colouring strength) data of saffron solutions (1% w/w) from the three countries were in the following order: IndiaIranUK. Also a good correlation (R2 0.84) was found between the chroma and colouring strength values. Similarly, mean aromatic strength values were the same order as colouring strength values, however, mean bitterness value of Iranian saffron was greater than that of Indian saffron followed by UK saffron. The release of safranal from aqueous solutions (1% w/w) of saffron under static headspace concentration showed that release of safranal from Indian saffron solutions was significantly (P<0.01) higher than that of Iran and UK saffron solutions, respectively. Overall, in terms of colour and flavour levels, there was a good correlation between the orders of origins of all three saffrons.
INTRODUCTION
Cultivation of saffron in new areas is one of the factors contributing to development of saffron either in quantity and/or quality. Saffron is known as an expensive crop as a result of labour costs and the extent of production worldwide however very little saffron creates intense flavour and strong colouring properties. Most important producing countries are Spain, Iran, Italy, Greece, India, Israel, China, France, Mexico and etc (Negbi, 1999). The main components of dried red stigmas of Crocus sativus L. (saffron) including crocin, picrocrocin and safranal are responsible for its colour, taste and odour respectively (Basker, 1999). Saffron stigmas when in natural form has red colour. By extracting with water, it produces a strong yellow colour called α-crocin (a water soluble caroteniod). Thus, the extractable colour intensity of saffron stigmas is determined using ISO/TS 3632-2 (2003) method as an International standard method. The bitter taste of saffron namely a glocoside picrocrocin is identified in commercial saffron using a standard method (ISO/TS 3632-2). Determination of safranal (odour component) is more problematic than other components since high flavour loss occur from production till packing of saffron stigmas and experimental procedure. The colour of saffron is of prime importance for consumers because of brilliant yellow food colour (Basker, 1999).
A colorimetric reflection method was used on saffron samples by Alonso and co-workers (2003). They concluded that the chromatic parameters (defined by CIE system L* (brightness), a* (redness-greenness) and b* (yellowness-blueness)) of saffron samples correlated well with their colouring power as well as it is a useful technique for quality assessment of saffron. Moreover, Hunter Lab system is similar to CIE system to measure colour of food product. Ferreira, Fernandz and Yotsuyanagi (1994) pointed out that the Hunter Lab and L*a*b* systems correlate well to sensory, visual and objective assessments of colour.Atmospheric chemical ionisation- mass spectrometry (APCI-MS) technique has been fully documented for quantitative analysis (Linforth & Taylor, 1993; Taylor, 1996; Taylor & Linforth, 2003; Taylor et al., 2000) in several review papers and reported as the reliable method for volatile flavour analysis in terms of simplicity and sensitivity (Benoit et al., 1983).
Aims of the present work were to evaluate saffron cultivated in the UK (England-East Midlands) and to compare it with two saffron samples (in filament forms without adulteration) from two producing countries Iran (Birjand) and India (Kashmir) in terms of three main components of saffron (colour, odour and taste).
MATERIALS AND METHODS
UK saffron was provided by the first author to be analyzed and compared with saffron samples obtained from Iran and Indiabought directly from producers. Safranal (C10H14O, 98-99% purity) was obtained from Sigma-Alderich (Gillingham, UK). Hexane as a solvent was supplied by Fisher Scientific (Loughborough, UK). The saffron stigmas were milled (their particle sizes: 0.5 mm through mesh sieve). They were placed insmall petri dishes (sealed containers) individually and covered with a polythene film and an aluminium foil to reduce any lossloses in their quality. Thereafter, they were kept in a freezer (-20 °C) before use for further experiments. Saffron stigmas (1g and three replicates of each sample) were placed in an oven (100°C) and keep for 16h. Thereafter samples were cool down in desiccators. After reach to room temperature, weigh the samples precisely. The moisture and volatile matter content, WMV, calculated as a percentage of the initial sample as follows:
WMV=% (Equation1)
Where m0 is the mass (in gram) of the sample and m4 is the mass (in gram) of dry residue. Each value of moisture and volatile content was averaged of three replicates of each saffron sample.
Characteristics of saffron- ISO/TS 3632-2:2003. The main characteristics of saffron were determined in terms of picrocrocin, safranal and crocin content using a UV/VIS spectrophotometer (Beckman DU-640, USA). By direct reading of absorbance, the bitterness (, maximum absorbance of picrocrocin), the aromatic strength (, maximum absorbance of safranal), and the colouring strength (, maximum absorbance of crocin) were obtained according to ISO/TS 3632-2. Each value of can be calculated as below:
(Equation 2)
Where D is a specific absorbance, m defined as mass of saffron sample in gram and H is moisture and volatile content of the sample, expressed as a mass fraction. A 1cm quartz cell was used as the sampling unit in the spectrophotometer. Each value was the average of three replicates.
Colour Measurement – Hunter Lab
Colour measurements of the samples were performed by using a Colourquest spectrophotometer SN C 5330 (Virginia, USA). The light source of D65 with observer angle of 10 was used. Colour parameters of L, a and b were taken in the Hunter Lab system. Saffron samples in powder or liquid forms were used in triplicates to obtain L, a and b parameters.The Hunter Lab color scale is more visually uniform than the XYZ color scale. In a uniform color scale, the differences between points plotted in the color space correspond to visual differences between the colors plotted. The Hunter Lab color space is organized in a cube form. The L axis runs from top to bottom. The maximum for L is 100, which would be a perfect reflecting diffuser. The minimum for L would be zero, which would be black. The a and b axes have no specific numerical limits. Positive a is red while a negative a is green. Positive b is yellow while a negative b is blue.
Atmospheric Pressure Chemical Ionization-Mass Spectrometry (APCI-MS)
A Platform quadrupole mass spectrometer (Micromass, Altrincham, UK) fitted with a custom built APCI interface was used to measure flavour release under static headspace. Water molecules entering the source of the MS are ionized by accepting protons (H+) and forming the hydronium ion (H3O+). The hydronium ion can then transfer its charge to any molecule with higher proton affinity. The molecular ions generated by APCI-MS can be differentiated due to ion mass/charge ratio (m/z value) thereby compounds with identical molecular weights and stereo isomers can not be discriminated.
Headspace measurements were performed using APCI-MS at a flow rate of 9 mL/min. Aliquots of the samples (10ml) were placed in 25ml flasks and were equilibrated at room temperature for 1h. Thereafter, the bottles were connected to the APCI-MS via the transfer line. The APCI-MS was operated in the positive ion mode. The safranal (151, m/z) compound was ionized by a 4kV corona discharge and a cone voltage of 20V in the source (50°C). The dwell time was 0.02s in the selected ion mode. The headspace concentration was obtained by peak height measurement of the signal.
The APCI-MS was calibrated by the introduction of hexane solution containing known quantity of Safranal (33.158mg), from which the concentration of the samples in headspace above sample determined. The concentration of safranal (33.158) is corresponding to a concentration of 100ppbv in the headspace when the volatile sampled at 75mL/min. The response from the samples was then converted into concentration (ppbv) in the gas phase using the relationship below:
Calibrant response= 100ppbvX 75/venture flow rate (equation 3)
RESULTS AND DISCUSSIONS
Saffron Characterization
Saffron samples from the UK, Iran and India countries were characterised based on the method specified in ISO/TS 3632-1 and 2, and were compared with values obtained from ISO/TS 332-1 as chemical requirements for saffron samples. The results (Table 1) showed that moisture content values of three countries sampleswere close (from 7.90-10.03) to the standard value (10), however the values of Indian saffron was slightly lower than others. This could be explained by using different drying procedure in different countries as well as climate differences.
The range of bitterness for standard values was 30-70. Only the bitterness value, 50.01 (for UK saffron) lied between the ranges, whereas other bitterness values (India, 75.55 and Iran, 76.15) were slightly higher than that of the maximum value in the bitterness range.The aromatic strength values, 33.34-44.72, of saffron from the three countries lie between the ranges for the standard values (20-50). It is interesting to note that the amount of safranal in all samples were approximately higher than the middle value confirming UK saffron can be produced for the quality purposes.
The colouring strength values were 65.01 (UK), 96.86 (Iran) and 115.83 (India). The two values from India and Iran lied between the ranges for colouring strength values (80-190) despite the UK value was lower than the minimum value in the range. It could be possibly explained by different crop production methods and ecophysiological climatic effects (e.g. the number of cloudy days in a year) and their interactions on growth and development of saffron in new regions.The coefficient of variation (CV) is defined as the standard deviation divided by the mean. The CV is a relative value and expressed as a percentage which can be used to compare the relative variabilities of two or more groups of values. Their CV values of all data presented in Table 1 were less than 10% which suggests low variability of data sets.
Colour Measurements
Colour of saffron is very important for consumers. The parameters L, a, b and c were obtained for colour determination. The L-value designates the lightness of the sample, where 100 value represents white and zero represents black. The b-value indicates redness when positive or greenness when negative. The b-value indicates yellowness when positive or blueness when negative. The c-value (chroma, (a2+b2)1/2) is a saturation index. Color parameters of saffron powder samples from UK, Iran and India were plotted in Fig. 1. It showed that the L-values were very close (50.07-53.53); however, other parameters were different. The a-value which indicates the redness increased from 12.76 (UK) to 19.91 (Iran). Alnso et al (2003) conclude that a reduction in a-value was in terms of increasing in whiteness as a result of style. The b-value which indicates yellowness increased from 10.20 (India) to 16.15 (Iran). The c-values were 16.61 (UK), 19.02 (India) and 25.64 (Iran). Thus, the color parameters of UK samples were almost minimum values, whereas, the parameters of Iranian samples were maximum values.
The parameters, L, a, b and c of saffron solutions (0.004%w/w) from UK, Iran and India were significantly different from those values of saffron powder samples (Figure 2). The L-values were quiet similar (96.33-98.48) for all three countries. The a-values which indicate the green color of saffron solutions increased from -8.92 (UK) and -8.15 (Iran) to -2.93 (India). The b-value, yellowness, decreased from 60.60 (UK) to 12.46 (India) as well as reduction of c-values from 61.26 (UK) to 12.80(India).
In terms of the redness of the powdered saffron (a value), the Indian saffron was the best followed by Iranian and UK saffrons. However in terms of the yellowness (b) of the saffron samples, the UK saffron had the highest value followed by Iranian and Indian samples respectively. The same trend can be seen for the saffron solutions (0.004%).The c-values (chroma) of powdered saffron of three countries obtained from HunterLab method were compared with colouring strength values due to ISO/TS 3632-2 method. The results showed that a good correlation (R2) of 0.84. It confirms that HunterLab method could be used as a standard method for saffron evaluation.
Headspace Analysis
The release of safranal from aqueous solutions (1% w/w) of saffron under static headspace concentration showed that release of safranal from Indian saffron solutions was significantly (P<0.01) higher than that of Iran and UK saffron solutions, respectively (Table 2). Their CV range was less than 20% (3.99-19.52%) which is relatively small for APCI-MS technique.
Aromatic Strength Values versus HS Concentration of Safranal
It is of interest for saffronologists to improve methodology to measure the quality of saffron samples due to simplicity and various choices. Thus it could help us to apply the best method in order to obtain reliable experimental results. In this regard, APCI-MS is a useful technique which is used extensively in studying flavour release. The aromatic strength values were compared with HS concentrations of safranal from saffron solutions (1%w/w). There was a good correlation (R2 of 0.68) between data from ISO/TS method and APCI-MS method. Therefore, APCI-MS could be a good substitute for determining aromatic strength values of saffron solutions. Nevertheless, the type of method is highly dependent on the accuracy required for data collection.
Conclusion
Colour properties of saffron are highly related to climate (light intensity and humidity) as well as processing condition for drying such as temperature and method of dehydration. Further research requires in order to producing better quality and more importantly growing saffron in the UK (England-East Midlands).Hunter Lab data presented appeared to be the most reproducible systems, with no significant differences between them for the illuminant and angle used. Hunter Lab provides useful information for quality control as its coordinates can serve to predict visual appearance. Therefore, the Hunter measurement system was the reliable method, compared with the other methods, for fast and objective colour evaluation. The results also suggested that APCI-MS could have some important applications in terms of instant monitoring of samples from raw materials to finished products. Ultimately, such methods could be of significant value in the analysis of saffron.
Literature Cited
Alonso, G.L., Sanchez-Fernandz, M.A., Seaz, J.R., Zalacian, A. and Salinas, M.R. 2003. Evaluation of the colour of Spanish saffron using tristimulus colorimetry, Italian J. Food Sci. 15:249-258.
Basker, D. 1999. Saffron Chemistry, In Saffron (Crocus sativus L.) (Ed. Ngbi, M.), Harward Academic Publishers, The Netherlands, pp.45-52.
Benoit, F.M., Davidson, W.R., Lovett, A.M., Nacson, S. and Ngo, A., 1983. Breath analysis by atmospheric pressure ionisation mass spectrometry, Analytic. Chem. 55:805-807.
Ferreira, V.L.P., Fernandz, V. and Yotsuyanagi, K. 1994. The color of chicken and pork meat loaf with added cured bovine blood as evaluated by RAB, HunterLab L-Asterisk A-Asterisk B-Asterisk and XYZ-CIE systems,Revista española de ciencia y tecnología de alimentos, Ferreira 34:311 -322.
Linforth, R.S.T. andTaylor, A.J. 1993. Measurement of volatile release in the mouth, Food Chem. 48:115-120.
Negb, M. 1999. Saffron cultivation: past, present and future prospects, In Saffron (Crocus sativus L.) (Ed. Ngbi, M.), Harward Academic Publishers, The Netherlands, pp.1-17.
Taylor, A.J. and Linforth R.S.T. 2003. Direct mass spectrometry of complex volatile and non-volatile flavour mixtures, Int. J. Mass Spect. 223:179-191.
Taylor, A.J. 1996. Volatile flavor release from foods during eating. Crit. Rev. Food Sci. Nut. 36:765-784.
Taylor, A.J., Linforth, R.S.T., Harvey, B.A. and Blake, B. 2000. Atmospheric pressure chemical ionisation mass spectrometry for in vivo analysis of volatile flavour release,Food Chem. 71:327-338.
Table 1. Characterization of saffron samples from the UK, Iran and India was obtained and compared with standard (ISO/TS 3632-2). The bitterness values at absorbance of 257 nm, the aromatic strength values at 330 nm and the colouring strength values at absorbance of 440nm were determined according to ISO 3632-1 and 2. Each value was calculated based on three replicates and their %CVs’ (%CV= Standard deviations*100/mean) were less than 10%.
Characteristic / standard / UK / India / Iran / Test methodMoisture and volatile matter, %(m/m) / 10 / 9.15 / 7.90 / 10.03 / ISO 3632-2
Bitterness (picrocrocn) / 30-70 / 50.01 / 75.55 / 76.15 / ISO 3632-2
Aromatic strength (safranal) / 20-50 / 33.34 / 44.03 / 40.72 / ISO 3632-2
Colouring strength (crocin) / 80-190 / 65.01 / 115.83 / 96.86 / ISO 3632-2
Country of origin / HS concentration (ppbv) / %CV
UK / 208 / 3.99
Iran / 864 / 19.52
India / 4472 / 16.38
Table 2. Headspace concentration (HS) of saffron solution (1%w/w) from UK, Iran and India and their coefficient of variations (%CV).
Fig. 1. Hunter Lab values including L, a, b and c of saffron powder samples obtaining from the UK, Iran and India. The error bars represent standard deviations of three replicates.
Fig. 2. Hunter Lab values including L, a, b and c for saffron solutions (0.004% w/w) of the three countries, UK, Iran and India. The error bars represent standard deviations of three replicates.