International Journal of
Fisheries and Aquaculture

  • Abbreviation: Int. J. Fish. Aquac.
  • Language: English
  • ISSN: 2006-9839
  • DOI: 10.5897/IJFA
  • Start Year: 2010
  • Published Articles: 216

Full Length Research Paper

Moisture sorption isotherm of Tuna Se’i, smoked with liquid smoke

Ayub U. I. Meko
  • Ayub U. I. Meko
  • Faculty of Fisheries and Marine Science, Sam Ratulangi University, Manado 95115, Indonesia.
  • Google Scholar
Siegfried Berhimpon
  • Siegfried Berhimpon
  • Faculty of Fisheries and Marine Science, Sam Ratulangi University, Manado 95115, Indonesia.
  • Google Scholar
I Ketut Suwetja
  • I Ketut Suwetja
  • Faculty of Fisheries and Marine Science, Sam Ratulangi University, Manado 95115, Indonesia.
  • Google Scholar
Frans G. Ijong
  • Frans G. Ijong
  • Faculty of Fisheries and Marine Science, Sam Ratulangi University, Manado 95115, Indonesia.
  • Google Scholar

  •  Received: 11 May 2016
  •  Accepted: 01 September 2016
  •  Published: 31 October 2016


One of the meat products in East Nusa Tenggara is called “se’i”, commonly made from beef, pork, deer meat, and recently have been developed using fish meat. Moisture Sorption Isotherm (MSI) is an important characteristic of food which is almost affected to aspect of the dehydration process and storage stability of the dried product. The aims of this study are to produce tuna se’i using liquid smoke with different concentrations, and to determine the MSIs using a constant relative humidity camber. In this method, water content of samples are allowed to equilibrate in a closed chamber with a salt solution which produces a desired relative humidity. Tuna se’i were made from two part of the tuna meat, that is, dorsal and belly. Five liquid smoke concentrations were used to dip the meats, that is, 1.0, 1.2, 1.4, 1.6 and 1.8%. The results show higher water conten in se’i from dorsal and the lower in se’i made from belly. The equilibrium moisture conents for samples  in lower RHs (< 54%)  ranged between 2.65 and 12.79% dry basis (db), while in higher relative humidities (RHs) (>54%) ranged  between 11.12 and 75.65% (db). Using five models equations, Oswin models shown best fit values of regression coefficients, with ranges value of relatife determination, 4.81 to 9.89, and MSI curves were sigmoid. From the MSI curves, can be conclude that tuna se’i need a packaging for storage in ambient temperature specially RHs above 50%.


Key words: Liquid smoke, se’i, moisture sorption isotherm (MSI), equilibrium moisture content.


Every region in Indonesia has a typical food preferred, in accordance with raw materials and additional materials are available locally. Se'i meat is one of the popular products processed in East Nusa Tenggara (ENT) especially community those living in mainland of West Timor. Since beginning, se'i are made from beef, buffalo, horse, deer and pork meats but recently, se’i were  also processed using fish  meat.  These  products  have  been widely introduced, especially through those who have been visited ENT.
Se'i  meat were processed by direct smoking using woods fire  at temperature of 60 to 100°C for 4 to 12 h, depend on meat size and ultimate  water content needed. There are dry product, for long shelf life, and half-dry product for short shelf life or for storage in refrigerator. Se’i   obtained   by   direct   smoking   usually    varied   in appearance, taste, and chemical composition. In addition, safety of products is still low. A modification of the processing using liquid smoke to dip the fish meat and then dried in cabinet drier, produced a good quality of se’i. The processing is more hygienic, and the product can be standardized (Berhimpon et al., 2014).
As a specific product, Tuna se'i have physical, chemical and microbiological characteristics, that is different from other food. One factor affecting the characteristics is a properties of moisture sorption isotherms (MSI) (Labuza, 1984).  Furthermore the general model of the food MSI curve can be used to design a model of handling and the shelf life of raw materials and  processed food  MSI  value could be influenced by the water content and water activity (aW) of each materials.
The presence of water in food can be classified in two types, that is, free water and bound water. The nature of free water in food is the same as an ordinary water which has a value of aW = 1, while the bound water is closely related to the composition of foodstuffs and have aW less than 1 (Labuza, 1984; Schmidt, 2007). Bound water contained in foodstuffs consists of three fractions namely primary bound water is also often called a monolayer of water, and secondary bound water, and tertiary bound water (Labuza and Altunakar, 2007), which generally serves as solvent for solid molecules, because it has a hexagonal structure (Soekarto and Adawiyah, 2012).
Foodstuffs in generally can absorb water from the surrounding (adsorption) because of the humidity outside material is higher, or vice versa can remove some of the water into the air (desorption) due to humidity outside material is low. The term sorption according to Van-den Berg and Bruin (1981) is the process by which solid of a food combines with water in a reversible process involving physical adsorption and capillary condensation. The equilibrium moisture content relationship with the activity of water at a certain temperature the sorption isotherm curve depicted on the water. Each food has a typical moisture sorption isotherm curve.
Labuza (1984) explains that there are three forms of MSI curve which is generally used to describe the moisture sorption models experienced by foodstuffs.  Furthermore, based on the state of water in foods that characterize aw of 0.0 to 1.0, then the MSI curve is divided into three parts. The first region with aW <3.0, second region with 3.0<aW<0.7 and third regions with aW>0.7 (Arslan and Togrul, 2005; Labuza and Altunakar, 2007).  Five mathematical models have been used in this research to fit the MSI data, that is, Hasley, Chen Clayton, Henderson, Caurie, Oswin.
The aims of this study are to obtain the equilibrium moisture content and mathematical model to fit  the MSI of  Tuna’s se'i. 


This research  was  conducted at the Fish Handling and Processing Laboratory, Faculty of Fisheries and Marine Sciences, Sam Ratulangi University, from October to December 2014. Materials used in this study are: Dorsal and belly meat of Tuna; Ingredients: salt, garlic, shallots, galangal, lemon grass, coriander and pepper. Smoking was done using liquid smoke produced by the laboratory (Berhimpon et al., 2014).  
Five liquid smoke concentrations were used to dip the meats, that is, 1.0, 1.2, 1.4, 1.6 and 1.8%.  Saturated salt solutions used to provide a range of nine types of RH, that is, NaOH, LiCl, MgCl2, K2CO3, Mg(NO3) 2, NaNO2, NaCl, KCl and K2SO4,  provided nine RHs of  6.9, 11, 32, 44, 53, 64, 76, 87 and 97%, respectively (Berhimpon, 1990; Chalermehat and Owasit, 2011; Ansari et al 2011).
The equipments used in processing of se'i namely: Cool box, knives, cutting boards, buckets, trays, glass bowl, and drying cabinets. While the device used for the determination of moisture sorption isotherm is, desiccators, sample bottles, porcelain bowls and analytical balance.
Fresh tuna were taken from Blue Ocean Grace International Company, at Aer Tembaga Bitung, Indonesia. Fillet were provided with a size of 12 × 2 × 2 cm and then washed, drained, weighed, and divided into 12 experimental units of 500 g each.  The fillets were soaked in the 2% salt solution for 30 min, drained for 20 min. Furthermore was the fillets divided into two groups. The first group soaked in five concentration liquid for 30 min, then drained for 20 min and soaked in spices solution for 2 h, then drained 20 min and dried in a cabinet dryer for 4 h at a temperature of 70 to 80°C.  The second group were followed a traditional method, were fillets  directly soaked  in spice solution  for two hours, then drained for 20 min and smoked  in smoking  cabinet  for 4 h at  temperature of 70 to 80°C (Meko, 2007).
To determine MSI, sample was crushed and then weighed about 2 g into the sample box.  Furthermore, the sample box was put into desiccators with different humidity. Weighing of the samples was done every 48 h until a constant weight was obtained (Rahman et al., 1995).  Equilibrium moisture content of samples that have reached constant weight were immediately measured using the oven method (AOAC, 2005).
Water sorption isotherm curve modeling
There are several mathematical models are commonly used in describing the MSI curve models of foodstuffs.  In this study, five models used to fix MSI curve was used (Table 1).
Appropriate model were selected by values of mean relative determinant (MRD). Formula (1) were used to calculate MRD:
Where: Mi = Levels of experimental results, MPi = water content calculation and n = number of data The value of MRD<5, the MSI models can describe the actual situation, the value of 5<MRD< 10, the model describes the state of approaching the truth, and MRD value >10, the model cannot describe the actual conditions (Isse et al., 1983; Bajpai et al., 2011).


Water content of Tuna’s Se'i
Water content of Tuna’s se’i made from dorsal meat is higher  than  that  of  Tuna’s  se’i  made  from  belly meat.
Concentration of liquid smoke used were not significant (P>0.05) on water content. Differences in water content can be seen in Figure 1. The figure show that the higher concentration of liquid smoke applied in Tuna se’I, the lower the water content. This is because the higher the concentration of liquid smokes the lower the pH.  The pH value of liquid smoke decreased with increased liquid smoke concentration, that is, 3.19, 3.13, 3.06, 2.95 and 2.85 for concentration of 1.0, 1.2, 1.4, 1.6, and 1.8 respectively. Low pH affected denaturation of meat and the denaturation can cause a decrease in the ability to hold water during the drying process. Soekarto and Adawiyah (2012) states that the food products or biological materials, have different ability to bind water,  depending on the type of chemical compound, the molecular structure, cell shape, pH and humidity of the environment. The water binding ability of meat and products is largely determined by pH’s isoelectric (Lawrie et al., 2003; Lonergan and Lonergan, 2005; Elvira, 2007). In addition to the effect of pH, protein denaturation is also caused by the heat during the drying process in cabinet dryer and smoking. Denaturation has a positive or negative effect on the water content of the meat (Zielbauer et al., 2016).
Tuna se’i produced, had moisture content ranges from 42.36 to 53.29% (wb) can be classified to intermediate food. Winarno (1993) stated that material which has a moisture content in the range of 15 to 50% (wb) is classified as an intermediate food. Meanwhile, according to Taukist and Richardson (2007), intermediate foods has a water content of 10 to 40% (wb) and aW  between 0.60 to 0.90.
Equilibrium moisture content
The equilibrium moisture content (EMC) of Tuna’s se'i made using liquid smoke  with  different  concentrations achieved during the 8 to 28 days after stored in RHs range from 7 to 97%. The higher the RHs the longer the EMC reached. Tuna’s se'i  samples were stored in RHs 7-32, 44-76 and 87-97%,  generally reached  equilibrium in average at 8-12, 14-20 and 24-28 days, respectively, with EMC average at 3.52-10.04, 11.37-20.29 and 34.09-66.83% (db) respectively.  
Labuza and Altunakar (2007) divided food into three groups, that is, monolayer, multy-layer, and free water layer.  If EMCs were grouped into the three groups, that is, mono layer (RHs <32%), multilayer (32-76% RHs) and free layer (RHs > 76%), the EMC of dorsal meat were <12.15, 12.15-22.61 and 22.61-75.7% (db) respectively, while the belly meat were <10.97, 10.97- 22.34 and 22.34-69.92% (db) respectively. This shows that the equilibrium moisture content in the dorsal meat are generally slightly higher than that in the belly. The different interval equilibrium moisture content due to the loin   is   striking   at   RHs>   76%,   because   the   moist environment allows the material to absorb more water vapor. Arslan and Togrul (2005) stated that at RHs>76% high humidity environments, allowing the material has a high equilibrium moisture content, because the water is trapped in the capillary spaces. Hysterical curve shape of food containing high protein such as soy, egg white and freeze-dried pork, show similarities with the lop hysterical wide with elevated levels of high water in aW 0.9 to 1.0 (Labuza and Altunakar, 2007; Catelam et al., 2011).
The relationship between RHs and the equilibrium moisture content describes the characteristic of the MSI. Moisture sorption isotherm of tuna se’i can be seen in Figures 2 and 3. The curves show a sigmoid, generally found in foodstuffs and products. The sigmoid curve is influenced by the accumulation of the combined effects of colligative, capillary and solid surface interaction with water, (Labuza and Altunakar, 2007; Bell and Labuza, 2000). The temperature and the composition of the material, can affect changes in the form  of  interaction  of water molecules in food (Wydiotomo et al., 2011; Catelam et al., 2011).  Iglesias and Chirife, (1976), noted that sorption region is generally influenced by the interaction of water molecules in food.
Both MSI curves of tuna se'i also shows that, in the range of RHs<76% equilibrium moisture content increased sharply, precisely at 87% RH. Berhimpon (1990) found that MSI of salted dried yellow tail have the same trend. Figures 2 and 3 explained that product will adsorb water when stored in ambient temperature with RHs > 75%, and in this condition packaging of the product is needed. The product been stored in RHs < 75% will be more stable.  This indicated that Tuna’s se'i at RHs > 87% has a high ability to absorb water from the environment. The interaction of muscle meat with moisture caused increase in tertiary bound waters. Soekarto and Adawiyah (2012) stated that in foodstuffs, organic compounds have a high ability to trap water in capillary. Furthermore they stated that the jelly and grass jelly are able to bind water 10 to 20 times of the dry weight while fresh fish 3 to 4 times, compared to grain is only 1/8 to 1/9 times the weight of the solid.
The accuracy of the models
Equilibrium moisture content were analyzed using the five equations model in Table 1, and the results were shown in Table 2.
Table 2 showed that there are two equation models (Hasley and Oswins models).with MRD value < 10. Hasley models only shown value >MRD 9.35 and 8.28 traditional se’i and se’i made of belly meat dip in 1% liquid smoke.  Chirief and Inlesias, (1978) stated that Oswin equation models have a high suitability for predicting MSI on foodstuffs with RHs of 0 to 85%.
Based on the value of MRD in Table 2, Oswin  models was chosen to fit MSI curve because it have a values of MRD tested <10 for all treatments. MSI curve is a relationship between aW, and the equilibrium moisture content of each treatment (Met) and the equilibrium moisture content prediction (Mep) (Bajpai et al., 2011).
Figure 4 shown that the curve was smoother; the lines have a high accuracy coincidence. The smaller the value of MRD, the more precise the model also describes the phenomenon in which the sorption isotherm occurs in foodstuffs and products (Walpole, 1990; Chalermachat and Owasit, 2011; Ansari et al., 2011). MSI curve smoothness in food is influenced by the form of equations and parameters used. Oswin equation model is a simple model of the five models selected.  A simple equation models with few parameters generally provide high curve smoothness (Labuza, 1982). Hadrich  et al. (2008) investigated the desorption of Tunisian sardine at temperatures  of  25, 35 and 50°C found Oswin equation  was the best  model to fit  the experimental or date,  whereas  Bajpai (2011) noted  that at temperature 10 and 70°C,  Gugenheim-Anderson-de Boer (GAB) and  Oswin models  give reliable results for meat and meat products.



The water content in the Tuna’s se’i made from dorsal meat was higher than belly meat. Decrease in water content is caused by an increase in liquid smoke concentration used. Equilibrium moisture content, reached varied between RHs <32, 32, 76 and >76%, that is 8, 12 and 24 days of storage, respectively. The MSI curve tuna se’i was a sigmoid shaped, Equation model that gives rather precise predictions was Oswin equation, with MRD value ranging from 4.81 to 9.67. The level of water absorption of Tuna’s se'i rose sharply at RHs >76%, therefore packaging of the product at the high RHs should be considered.


The authors have not declared any conflict of interests. 


The authors would like to thank the United Board for Higher Education in Asia (UBCHEA) that funded the research and publication.


[AOAC] Association of Official Analytical Chemists (2005). Official Methods Of Analysis. Assoc. Office Anal. Chem. Washigton DC.


Ansari S, Farahnaky A, Majzoobi M, Baddi F (2011). Modeling the effect of glucose syrup on the moisture sorption of figs. J. Food Biophys. 6:377-389.


Arslan N, Togˇrul H (2005). Modelling of water sorption isotherms of macaroni stored in a chamber under controlled humidity and thermodynamic approach. J. Food Eng., 69(2):133-145.


Bajpai SK, Chand N, Lodhi R (2011). Moisture Sorption isotherms and heat of sorption of sago starch films. Asia J. Food Agro-Ind. 



Bell LN, Labuza TP (2000). Moisture sorption: Practical aspect of isoterm. Am. Assoc. Cereal Chemist. Minesota, USA.


Berhimpon, S., Souness, R.A., Buckle, K.A. and Edwards, R.A., 1990. Salting and drying of yellowtail (Trachurus mccullochi Nichols). International Journal of Food Sci. Technol., 25(4):409-419.


Berhimpon S, Montolalu IR, Dien AH, Mentang F (2014). Devalopment of Exotic Smoked Fisf with Low Carcinogenic Using Waste of Fishery Industry. Report Indonesian Ministry of Risert and Higer Education. Sam ratulangi University, Manado.


Catelam KT, Trindade CSF, Romero JT (2011). Water adsorption isotherms and isosteric sorption heat of spray-dried and freeze-dried dehydrated passion fruit pulp with additives and skimmed milk. Ciênc. e Agrotecnol. 35(6):1196-1203.


Chalermachat Y, Owasit P (2011). Water adsorption Isotherms and thermodynamic analysis of Thai style marinated dried fish. (Pla Sawan).


Chirief J, Iglesias HA (1978). Equation for fitting water sorption isotherm of food. J. Food Tech. 13:159-165.


Elvira S (2007). Effect of pH on Meat and Dairy. 

View. Accessed 5 Mare 2016.


Hadrich B, Boudhrioua N, Kechaou N (2008). Experimental and mathematical study of desorption isotherms of Tunisian Sardine (Sardinella aurita). Food Bioprod. Process 86(4):242-247.


Iglesias HA, Chirife J (1976). Prediction of the effect of temperature on water sorption isotherms of food material. Int. J. Food Sci. Technol. 11(2):109-116.


Isse MG, Schuchmann H, Schubert H (1993). Divided sorption isotherm concept: an alternative way to describe sorption isotherm data. J. food process eng., 16(2):147-157.


Labuza TP (1982). Shelf life dating of foods. Food and Nutritions. Press Inc.


Labuza TP (1984). Moisture sorption: practical aspect of isotherm measurement and use. American Association Chemists. St. Paul. Minnesota.


Labuza TP, Altunakar B (2007). Water activity prediction and moisture sorption isotherms. In Canovas GVB, Fanata JA, Schmidt SJ, Theodore TP (Eds.). Water activity in food. Fundamentals and applications 1st edn IFT Press. Blackwell Publishing. pp. 109-154.


Lawrie RA (2003). Meat Science. Translator A. Parakkasi. Fifth ed. UI-Press.


Huff-Lonergan E, Lonergan SM (2005). Mechanisms of water-holding capacity of meat: The role of postmortem biochemical and structural changes. Meat Sci. 71(1):194-204.


Meko AUI (2007). Study of Temperature Difference and Long Smoking Effect to Se'i Tuna Organoleptic Characteristics. Int. Res. Report. Artha Wacana Christian University. Kupang.


Rahayu WP, Arpah M, Diah E (2010). Shelf Life Prediction and Isotherm Sorption Model of Dried Grain and Powdered Black Pepper (Piper ningrum L.). J. Teknologi Dan Industri Pangan 16(1):31.


Rahman MS, Sablani SS, Al–Ruzeiqi MH, Guizani N (2002). Water adsorption isotherms of freeze–dried tuna meat. Transactions of the ASAE 45(3):767.


Schmidt SJ (2007). 4 Water Mobility in Foods. Water Activity in Foods, P 47.


Soekarto ST, Adawiyah DR (2012). Keterkaitan berbagai konsep interaksi air dalam produk pangan [Interrelation on Water Interaction Concepts in Foods]. J. Teknologi Dan Industri Pangan, 23(1):107.


Taukist PS, Richardson M (2007). Principel of intermediate moisture food and related technology. In Barbaso-Canovas GV, Fontana JA, Schmidt SJ, Theodore TP (Ed). Water activity in food. Fundamentals and applications 1st edn IFT Press. Blackwell Publishing. pp. 273-312.


Van-den- Berg C, Bruin S (1981). Water activity its estimation in food systems: Theoretical aspect.


Walpole RE (1990). Introduction to statistics. Gramedia Pustaka Utama, Jakarta.


Widyotomo S, Atmawinata O, Purwadaria HD (2011). Water sorption isotherm characterization of green coffe beans by BET and GAB Models. Article Agritech. 



Winarno FG (1993). Nutritive Food, Technology and Consume. Gramedia Pustaka Utama, Jakarta.


Zielbauer BI, Franz J, Viezens B, Vilgis A (2016). Physical Aspects of Meat Cooking: Time dependent thermal protein denaturation and water loss. J. Food Biophys. 11:34-42.