Application of SeDeM Expert system in formulation development of effervescent tablets by direct compression

Abstract

The SeDeM expert system is a pre formulation tool applied for the prediction of the suitability of a material for direct compression. This innovative tool provides an index of good compressibility of the material indicating its aptitude to be compressed by direct compression. In the study the SeDeM expert system has been applied for the prediction of the behavior of the material to be used in the formulation of effervescent tablets by direct compression. Different formulations were developed on the basis of the results of the SeDeM expert system. Various parameters for the material as per the SeDeM expert system were determined according to their official and reported methods. Powder blend for different formulations was evaluated for their rheological properties while tablets were evaluated for various official and unofficial tests.

Suitability of the material for direct compression was successfully predicted using the SeDeM expert system. Domperidone was found unsuitable for direct compression. During formulation all excipients responded as they were predicted as per the SeDeM expert system. Tablets produced using the resultant formulations were having sufficient mechanical strength, free of premature effervescence and were capable to be scaled up for commercial manufacturing.

1. Introduction

SeDeM expert system is a pre formulation methodology applied for the formulation development of solid dosage form (Tablets) by direct compression (Pilar et al., 2006). Quality by design ICH-Q8 provides a basis for the SeDeM expert system. It is used for an evaluation of critical quality attributes having an impact on the final product. This system provides a physical profile of A.P.I. and excipients intended to be used and suggests their suitability for direct compression (Johnny et al., 2009). It also points out parameters needing to be improved to get a formulation that can be successfully processed by direct compression i.e., the profile of powder shows its advantages and gaps for their suitability for direct compression (Pilar et al., 2006; Inderbir and Pradeep, 2012).

The SeDeM expert system also calculates the amount of excipients with certain characteristics required for correction of a particular property in order to make the final blend suitable for direct compression. Several parameters have been selected that must be fulfilled by the formulation to ensure successful and robust processing by direct compression technology (Johnny et al., 2012; Josep et al., 2008).

Effervescent tablets are a promising dosage form combining qualities of both solid and liquid dosage forms. These are dissolved or dispersed in water before administration and taken as liquid thus presenting the drug in a palatable liquid form while retaining the properties of a solid dosage form like easy portability, high stability and accurate dose (British Pharmacopoeias, 2008). pH of the liquid formed after effervescence reaction can be controlled in the desired range by a proper selection of the quantities of acids and base. Furthermore, as the drug is administered as a liquid, the whole of it is made available for absorption from GIT (Ashutosh et al., 2008; Harald, 2003).

The main problem with effervescent tablets is their chemical instability exhibited by the premature effervescent reaction. Even trace amounts of the water can initiate the self propagating reaction that continues till the consumption of the whole of the acid and/or base resulting in a complete deterioration of the product (Harald, 2003). Therefore the process of preparation should be carried out in an environment of controlled humidity with a reduced number of steps to minimize material exposure. The method of direct compression is desirable for the preparation of effervescent tablets (Robert, 2001; Yuhua and Diana, 2009) as it involves fewer steps and less material handling and exposure (Harald, 2003). Main problem with the direct compression method is the prediction of material flow and compressibility. Most of the APIs lack sufficient flow and compressibility and requires selecting proper excipients for their formulation by direct compression. A large number of trials should be carried out to obtain formulations with proper rheological properties and compressibility. This makes the process more tedious, time consuming and a lot of material is utilized. The SeDeM expert system can overcome the problem as it develops a database for excipients and an easy selection can be made without extra trials.

The SeDeM expert system has been applied for the prediction of the suitability of different material used in the formulation of effervescent tablets by direct compression. The results predicted by the SeDeM expert system have been confirmed by an analysis of trials of the different formulations.

Various formulations were developed containing the effervescent pair alone and in combination with super disintegrants. Effects of super disintegrant, tablet compression force and tablet surface area have been evaluated.

Domperidone (5-chloro-1-h1-(3-(2,3-dihydro-2-oxo-1H-benzimidazol-1-yl) propyl)-4-piperidinyl-1,3-dihydro-2H-benzimidazol-2-one) was selected as a model drug. It is a dopamine-receptor antagonist acting peripherally, having no central effects with the elimination half life of 5–7 h (David et al., 1998). According to bio pharmaceutical classification system, domperidone has been classified as a class 2 drug. It is a weak base having a good solubility at lower pH (Nagarsenker et al., 2000). It absorbs well when the whole of the drug is available for absorption in the acidic segment of G.I.T. This can be made possible by administering domperidone as an effervescent tablet.

2. Materials and methods

2.1. Material

Domperidone (Ningbo Sansheng Pharmaceuticals Company, China) was purchased from Medicraft pharmaceuticals, Peshawar. Citric acid anhydrous, tartaric acid and sodium bicarbonate (Merck KGA, Germany) were purchased from sigma chemicals, Karachi. Rest of the excipients (Micro crystalline cellulose (F.M.C. Bio polymers, Ireland), Tablettose (Molkerei Meggle, Germany) and Magnesium stearate (Peter Greven, Malaysia) were a kind gift from Ferozsons Laboratories, Ltd., Nowshera. All the materials were of pharmaceutical grade.

2.2. Methods

2.2.1. Evaluation of material as per SeDeM expert system

Powder material was evaluated for different parameters according to the SeDeM expert system to determine their suitability for direct compression. Some of them were determined experimentally according to the established procedure and some were calculated from experimental values of other parameters (Pilar et al., 2006) as per Table 1.

Table 1.

All parameters of SeDeM expert system.

Factor/incidence	Parameter	Symbol	Unit	Equation	Limits	Applied factor
Dimension	Bulk density	Da	g/ml	Da = P/Va	0–1	10 V
	Tapped density	Dc	g/ml	Dc = P/Vc	0–1	10 V
Compressibility	Inter particle porosity	Ie	0	Dc–Da/Dc × Da	0–1.2	10 V/1.2
	Carr’ index	Ic	%	100x(Dc–Da)/Dc	0–50	V/5
	Cohesion Index	Icd	N	Experimental	0–200	V/20
Flow ability/powder flow	Hausner ratio	IH	–	Dc/Da	3–1	(30–10 V)/2
	Angle of repose	(α)	0	tan−1(h/r)	0–50	10–(V/5)
	Powder flow	t″	S	Experimental	0–20	10–(V/2)
Lubricity/stability	Loss on drying	%HR	%	Experimental	0–10	10–V
	Hygroscopicity	%H	%	Experimental	0–20	10–(V/2)
Lubricity/dosage	Particles < 50	%Pf	%	Experimental	0–50	10–(V/5)
	Homogeneity index	Iθ	–	Fm/100 + ΔFmn	0–2 × 10−2	500 V

2.2.2. Determination of basic parameters

The basic parameters of the SeDeM expert system are given as;

• •Bulk density
• •Tapped density
• •Interparticle porosity
• •Carr’s index
• •Cohesion index
• •Hausner ratio
• •Angle of repose
• •Powder flow
• •Loss on drying
• •Hygroscopicity
• •Particle size smaller than 50 μm
• •Homogeneity index

2.2.3. Conversion of experimental values (V) to radius value (r) of SeDeM diagram

The numerical values for different parameters of the material obtained by experimental determination were converted into a radius value (r) of the SeDeM expert system diagram. For the conversion of experimental value of each parameter, specific factors were applied (Johnny et al., 2009) as listed in Table 1.

2.2.4. Graphical presentation of SeDeM diagram

SeDeM diagram was built up on the basis of 12 parameters looking as 12 sided polygon (Johnny et al., 2009). Results obtained from the experimental determination of various parameters were converted and presented as a SeDeM diagram as shown in Fig. 1.

Figure 1.

Blank SeDeM diagram. Da, bulk density; Dc, tapped density; Ie, inter-particle porosity; IC, Carr index; ICd, cohesion index; IH, Hausner ratio; α, angle of repose; t″, flow ability; %HR, loss on drying; %H, Hygroscopicity; %Pf, Particle size; Iθ, Homogeneity index.

2.2.5. Calculation of I.G.C.

For determination of suitability of the material for direct compression the following indices are calculated on the basis of the SeDeM system as below (Johnny et al., 2012).

2.2.6. Parameter index

$$\text{I.P.}=\frac{\text{No. P}⩾5}{\text{No. Pt}}$$ (1)

where

• No. P ⩾ 5 = parameters with values equal to or more than 5.

• No. Pt = Total number of parameters.

Acceptability limit corresponds to a score of 5.

2.2.7. Parameter profile index

• I.P.P. = Average of r value of all parameters

The acceptable limit corresponds to a score of 5.

2.2.8. Good Compressibility Index

$$\text{I.G.C.}=\text{I.P.P.}\times f$$ (2)

where

• f = reliability factor.

2.2.9. Selection of acid to base ratio

Acid to base ratio of the effervescent pair was determined on the basis of stichometric calculations of balanced acid and base neutralization reaction. Calculated amount of acid and base were added to 250 ml of water having pH 7 at ambient temperature. After completion of acid base reaction, the pH of the solution was determined to observe any remaining acid or base.

2.2.10. Determination of per tablet quantity of effervescent pair

Placebo tablets were prepared with the same excipients and same compression weight having varying degrees of effervescent components. Three levels were studied for both acids separately. These were:

• 10% w/w

• 20% w/w

• 30% w/w

Compressed placebo tablets were subjected to an evaluation for effervescence time using 250 ml of purified water with pH 7 at ambient temperature. An average of 6 determinations of effervescence time was taken as the effervescence time of the combination. Results were presented as average ± S.D.

2.2.11. Taste evaluation of tablets

For taste evaluation placebo tablets were prepared containing taste making agent in different concentrations (1% w/w, 2% w/w, 3% w/w, 4% w/w and 5% w/w) as shown in Table 2. Taste making agent consisted of a fixed quantity of flavor (tutti fruity 0.5%) and sweetener (aspartame) making the remaining quantity. Tablets were dispersed and their taste was evaluated by a panel of 8 healthy human volunteers (30–45 years) having one hour as the washout time between two determinations. Observations of each volunteer were recorded on a scale ranging from tasteless to better tasting.

Table 2.

Formulations of placebo tablets for taste evaluation.

Ingredients	TEE-01	TEE-02	TEE-03	TEE-04	TEE-05	TEE-06
Micro crystalline cellulose	22	22	22	22	22	22
Tablettose-80	51	49	48	48.5	47.5	46.5
Citric acid anhydrous	10	10	10	10	10	10
Sodium bicarbonate	10	10	10	10	10	10
Flavor	0	0.5	0.5	0.5	0.5	0.5
Aspartame	0	1	2	3	4	5
Colloidal silicon dioxide	1	1	1	1	1	1
Magnesium stearate	1.5	1.5	1.5	1.5	1.5	1.5
C.C. Sodium	2.5	2.5	2.5	2.5	2.5	2.5
S.S. Glycolate	2.5	2.5	2.5	2.5	2.5	2.5

Quantities are given as % w/w.

2.2.12. Tablet preparation

The method of direct compression was applied for tablet preparation. All materials were weighed accurately according to their respective formulations as presented in Table 3. They were sifted through a stainless steel mesh with a pore size of 0.85 mm (Endicott, England) and were blended in a laboratory scale double cone mixer for 15 min at 20 rpm.

Table 3.

Formulations of effervescent domperidone tablets.

Ingredients	ED-01	ED-02	ED-03	ED-04	ED-05	ED-06	ED-07	ED-08	ED-09	ED-10	ED-11	ED-12
Domperidone	1.67	1.67	1.67	1.67	1.67	1.67	1.67	1.67	1.67	1.67	1.67	1.67
Citric acid	10	10	10	10	10	10	0	0	0	0	0	0
S.B.C.	10	10	10	10	10	10	10	10	10	10	10	10
Tartaric acid	0	0	0	0	0	0	10	10	10	10	10	10
S.S. Glycolate	0	0	0	3	5	2.5	0	0	0	3	5	2.5
C.C.Na	0	3	5	0	0	2.5	0	3	5	0	0	2.5
Mg. Stearate	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5
Flavor	3	3	3	3	3	3	3	3	3	3	3	3
M.C.C.	15	15	15	15	15	15	15	15	15	15	15	15
Tablettose	58.83	55.83	53.83	55.83	53.83	53.83	58.83	55.83	53.83	55.83	53.83	53.83

Quantities are given as % w/w.

S.B.C, Sodium bicarbonate; S.S. Glycolate, Sodium starch glycolate; C.C.Na, Cross carmellose sodium (cross linked carboxy methyl cellulose sodium); Mg. Stearate; Magnesium stearate; M.C.C, Micro crystalline cellulose.

Tablets were prepared by compressing powder blend using rotary compression machine D3A (Manesty, England) fitted with round flat 13 mm round punches having bisection line. Theoretical weight per tablet was 600 mg and 200 tablets were compressed for each formulation. The whole of the processing was carried out under the controlled conditions of humidity (relative humidity below 35%).

2.2.13. Evaluation of powder blend

Prior to compression, the powder blend was evaluated for their flow and compressibility. Various parameters related to the flow and compressibility like bulk density, tapped density, angle of repose, flow ability, compressibility index and Hausner ratio were determined according to official methods. All determinations were made in triplicate and results were presented as average ± S.D.

2.2.14. Tablet evaluation

Tablets from each formulation were evaluated (British Pharmacopoeias, 2008) for various official and unofficial parameters as under.

2.2.15. Physical properties of tablets

Weight variation was calculated after measuring 20 tablets individually (United States pharmacopoeia (USP 31 NF-26) and revision, 2008) using a digital balance (Libror AEG-120, Schimadzu, Japan). The deviation of individual weight from average weight was calculated and weight variation was calculated.

The thickness of 10 tablets from each formulation was measured using a digital tablet tester (Pharma test, Hamburg) and their average and standard deviation were calculated.

2.2.16. Mechanical strength of tablets

Crushing strength of 10 tablets, from each formulation, was determined using a digital tablet hardness and thickness tester (Pharma test, Hamburg) and their mean values were calculated. From mean values of crushing strength, diameter thickness, and tensile strength of tablets and specific crushing strength of tablets were calculated (Chuan-Yu et al., 2005; De Jong, 1987).

Friability was determined according to official monograph (British Pharmacopoeias, 2008) using single drum friabilator (Faisal Engineering, Pakistan).

2.2.17. Drug content

Domperidone content of tablets was determined as per the official method described in B.P., 2009 (British Pharmacopoeias, 2008) using methanol as solvent and as blank solution.

Absorbance of the sample solution and standard solution was measured at 284 nm using a double beam UV–visible spectrophotometer (Perkin Elmer, USA) (British Pharmacopoeias, 2008). The drug content was calculated by a comparison of absorbance of two solutions. All determinations were made in triplicate and their average and standard deviation were calculated.

2.2.18. Effervescence time of tablets (disintegration time)

Effervescence time was determined as per European pharmacopeia, by allowing one tablet to disperse completely in 250 ml of purified water at room temperature (European pharmacopoeia et al., 2005). Time required for the completion of effervescence was noted using a digital stopwatch (Sony, Japan). Effervescence time determination was performed for 6 tablets and results were presented as average ± S.D.

2.2.19. Moisture content

The moisture content of the tablet was determined using a Halogen moisture analyzer (Mettler Toledo, Switzerland). Powdered tablets were loaded to the pan of moisture analyzer and loss on drying was recorded. Determination was made in triplicate and results were presented as average ± S.D.

2.2.20. Wetting time of tablets

Wetting time of the tablets from each formulation was determined by placing the tablet on a filter paper soaked in a watch glass containing 5 ml purified water. The time required for complete hydration of the tablet was noted with a digital stopwatch (Sony, Japan). The experiment was performed in triplicate for each formulation and average waiting time was calculated (Zade et al., 2009).

2.2.21. Study of effect of different parameters on rate of effervescence reaction

To study the effect of different parameters on effervescence time, tablets were prepared from optimized formulation by varying these parameters as under;

2.2.22. Effect of surface area of the tablet

To study the effect of surface area of the tablet on effervescence time, all formulations were compressed on smaller punches (10 mm, oval) and larger punches (13 mm round) and their effervescence time was compared. To minimize the effect of compressibility on effervescence time, hardness of tablets compressed on smaller punches was such that its tensile strength and specific hardness were comparable to those of larger sized tablets.

2.2.23. Effect of super disintegrants

Two super disintegrants (cross linked carboxy methyl cellulose sodium and sodium starch glycolate) were used to study their effects on effervescence time of the tablets. Effervescent tablets were prepared with 20% of effervescent pair and at a hardness level of 7–10 kg. Disintegrants were added alone (concentration levels 3% and 5%) and in combination (2.5% w/w each) and their effervescence time was compared.

2.2.24. Effect of tablet compression force

To study the effect of compressibility on the effervescence time of the tablets, tablets from optimized formulation were compressed at 3 levels of hardness i.e.

• 4–7 kg

• 7–12 kg

• 12–16 kg

Effervescence time was determined for 6 tablets, from each level, using 250 ml of purified water, their average was calculated and compared with each other. Results were presented as average ± S.D.

3. Results and discussion

3.1. Evaluation of material as per SeDeM expert system

Domperidone is a white non flowing powder and was evaluated for 12 parameters as per the SeDeM expert system. As evident from the data given in Table 4, dimension, compressibility, and flow ability/powder flow factors for domperidone are lower than acceptable values and need to be improved in order to get tablets by direct compression. Index of good compressibility was calculated for domperidone and was found to be well below the acceptable limit of 5 (Table 5).

Table 4.

Evaluation of domperidone as per SeDeM expert system.

Parameter	Results
	V	“r”
Bulk density	0.226	2.26
Tapped density	0.336	3.36
Inter particle porosity	1.45	0
Carr’ index	32.74	6.548
Cohesion Index	66	3.3
Hausner ratio	1.487	7.565
Angle of repose	40	2
Powder flow	0	0
Loss on drying	2.34	7.66
Hygroscopicity	6.39	6.81
Particles < 50	1	9.8
Homogeneity Index	0.014	5.2

V, Experimental values; “r”, Converted experimental value.

Table 5.

“r” Values of domperidone and excipients as per SeDeM expert system.

Parameter	Diluents		Effervescent material
	M.C. cellulose	Tabletose	C. acid	T. acid	S.B.C
Bulk density	3.85	6.1	7.52	9.21	6.8
Tapped density	5.26	7.38	9.17	10	10
Inter particle porosity	5.8	4.33	1.99	0.87	3.92
Carr’ index	5.36	3.47	3.6	1.923	6.4
Cohesion Index	5.65	6.3	5.7	6.6	4.45
Hausner ratio	8.2	8.95	8.9	9.45	7.65
Angle of repose	4.2	5.6	5.8	5.2	3.8
Powder flow	6	7	6.5	6	4
Loss on drying	5.93	9.52	6.28	5.7	7.84
Hygroscopicity	8.29	9.055	8.61	8.395	8.595
Particles < 50	7.41	9.93	9.9	9.86	9.72
Homogeniety Index	6.42	6.3	5.3	5	5.6

Results are presented as “r” value of SeDeM diagram.

M.C. cellulose, micro crystalline cellulose; C. acid, citric acid; T. acid, tartaric acid; S.B.C., sodium bicarbonate.

Materials with better flow and compressibility are required for preparation of their effervescent tablets by direct compression. Diluents and effervescent pair are the 2 main excipients that will play their role to overcome deficiencies of domperidone and making it suitable for direct compression. So diluents with higher values of the above mentioned factors should be selected for the formulation of effervescent domperidone tablets.

As evident from the results presented in Table 5 and SeDeM diagram for micro crystalline cellulose (Fig. 2), most of the studied parameters of MCC were within the normal range, i.e. above 5. Only 2 parameters bulk density and angle of repose had “r” values below 5 which resulted in an average “r” value of compressibility factor below the limit.

Figure 2.

SeDeM diagrams of domperidone and main excipients.

“r” Value of angle of repose is also below 5 indicating its poor flow ability. The flow of MCC will increase with the addition of lubricants and colloidal silicon dioxide into the formulation. The results of the rest of the parameters prove its suitability for direct compression.

Comparison of SeDeM profile of 2 diluents shows that they are suitable for the preparation of ODTs by direct compression. (Fig. 3) MCC has a lower “r” value of bulk density and angle of repose while tablettose has both of the parameters on the upper side and expected to cover the shortage. Similarly tablettose is deficient in both parameters of compressibility which are very high for MCC and will compensate lower values of tablettose-80. The rest of the parameters for both are well above 5 proving that they can be successfully applied to direct compression. The I.G.C. value of both the diluents was calculated and found to be well above the acceptable limit. In combination they will provide an efficient diluent system for the formulation of ODT by direct compression. On the basis of SeDeM results of the micro crystalline cellulose and tablettose (Fig. 3) they will be used in combination as diluents. They will improve the flow and compressibility of domperidone to the extent of making it suitable for direct compression.

Figure 3.

Comparison of tablettose and MCC.

Citric acid is a crystalline solid (Handbook of Pharmaceutical Excipients and Edition, 2009). To get uniform size powder and increase its surface area, it was pulverized through mesh number 40 and evaluated for different parameters as per the SeDeM expert system. As evident from the data presented in Table 5. Average “r” values for all the factors were within the normal range of 5–10. Only compressibility factor has average “r” values lower than 5 indicating its poor compressibility. Two of the parameters (inter particle porosity and Carr’s index) included in the factor have “r” values below the limit. The I.G.C. was calculated for citric acid and was found to be above 5 showing its suitability for use in direct compression.

Like citric acid, tartaric acid was also pulverized through mesh number 40. Both citric acid and tartaric acid have almost similar results. Tartaric acid also had a lower compressibility factor and the rest of the factor was within the acceptable range of 5–10. The I.G.C. value of tartaric acid was calculated to be 6.204. It was above the acceptable limit showing its suitability for direct compression.

Sodium bicarbonate has very poor rheological properties as evident from the data presented in Table 5. Although its I.G.C. value was within the acceptable range (Table 6), still it cannot be used alone in direct compression requiring a large quantity of excipients with good flow and compressibility for compensation. Its I.G.C. is high due to the high value of some of the parameters. Some of its parameters have an “r” value at the uppermost side (close to 10) while some have values very close to zero.

Table 6.

Various indices of material as per SeDeM expert system.

Ingredient	I.P.	I.P.P.	I.G.C.
Domperidone	0.5	4.542	4.324
Micro crystalline cellulose	0.833	6.114	5.821
Tablettose-80	0.833	6.995	6.659
Cross carmellose sodium	0.75	6.319	6.016
Sodium starch glycolate	0.667	6.04	5.75
Citric acid	0.833	6.606	6.289
Tartaric acid	0.833	6.157	6.204
Sodium bicarbonate	0.667	6.565	6.25

I.P, parameter index; I.P.P, parameter index profile; I.G.C, index of good compressibility.

3.2. Selection of acid to base ratio

Both the acidic moieties (citric acid and tartaric acid) intended to be used in the formulation of effervescent domperidone tablets were pulverized through mesh number 40 using a rotary granulator (STC, China). After pulverization they were dried at 45 ± 5 °C for 1 h to remove absorbed moisture.

Sodium bicarbonate was heated at 120 °C for 30 min. At elevated temperatures some of sodium bicarbonate is converted into sodium carbonate forming a protective layer and the surface of sodium bicarbonate gets passive. The surface passive sodium bicarbonate was used in the formulation of effervescent tablets.

Quantity of acid and base for effervescence reaction was calculated on a molar basis of their balanced psychometric equation. They were allowed to react in purified water and the pH of the solution was noted for any remaining acid or base. The pH of the resultant solution was on the acidic side showing complete consumption of sodium bicarbonate. The unreacted citric acid resulted in the acidic pH of the solution which improved taste perception.

In the case of tartaric acid the pH of the solution was the alkaline showing the presence of free sodium bicarbonate in the solution. Tartaric acid is highly hygroscopic and absorbs atmospheric moisture to a greater extent. The high moisture content reduces tartaric acid quantity per weight. On that basis tartaric acid was dried properly before inclusion into the formulation and its quantity was also slightly more than that required for neutralization of sodium bicarbonate.

3.3. Determination of effervescent pair quantity per tablet

As shown in Table 7, very small disintegration time can be achieved with effervescent material constituting 30% w/w of the total tablet weight (57.83 ± 3.06 s). Still effervescent material is not included in this concentration because;

• •Sodium bicarbonate has very poor compressibility and rheological properties [20]. Its high concentration can affect the final product adversely.
• •The difference between disintegration time with 20% w/w and 30% w/w effervescent material is very low.

Table 7.

Disintegration (effervescence) time v/s quantity of effervescent material.

Qty of acid/base pair (%)	ET-01	ET-02	ET-03	ET-04	ET-05	ET-06	Avg. E.T.	Std. Dev.
10	137	141	132	139	134	129	135.33	4.502
20	68	77	71	64	74	69	70.5	4.594
30	59	56	54	61	53	58	56.833	3.061

E.T., effervescence time (Second); Avg. E.T., average effervescence time; Std. Dev., standard deviation.

Due to these reasons effervescent material was included in the formulation of effervescent tablets in 20% w/w of the total weight of the tablet. Same ratio was applied for citric acid/sodium bicarbonate and tartaric acid/sodium bicarbonate pairs.

3.4. Taste evaluation

The taste making agent combination was composed of a sweetening agent (aspartame) and flavor (tutti fruity). As evident from the volunteer’s response (Table 8) formulation without any taste making agent (sweetener and flavor) had an acceptable taste because domperidone is a tasteless material. The taste of the other formulation ingredients was dominated by an acidic taste of citric acid. Quantity of citric acid was slightly more than that required for neutralization of sodium bicarbonate on the basis of stoichiometric calculations. On the basis of the volunteer’s response, the taste making agent was included into the formulation at the level of 3% as at this concentration the taste was marked as pleasantly sweet. A concentration of 4% and 5% of the taste making agent was ranked as strongly sweet by most of the volunteers.

Table 8.

Volunteers’ response about taste.

Formulation	Number of volunteers rated tablets as
	0	1	2	3	4
TEF-01	–	8	–	–	–
TEF-02	–	3	4	1	–
TEF-03	–	–	7	1	–
TEF-04	–	–	1	7	–
TEF-05	–	–	–	3	5
TEF-06	–	–	–	–	8

0, Bitter tasting; 1, Acceptable; 2, Pleasant; 3, Sweet; 4, Strongly sweet.

3.5. Evaluation of powder blend

Tablettose-80 and micro crystalline cellulose were collectively used as diluents on the basis of their SeDeM profile. Due to their good flow and compressibility both of these excipients were able to compensate the poor flow and compressibility of domperidone and other excipients in the formulation (e.g. Sodium bicarbonate). It is evident from Table 9, that all formulations have a very good Hausner ratio, Carr index and angle of response. Angle of repose for all formulations was less than 31°. ED-12 showed the angle of repose to be worst of all formulations even then it was on the good side. There was a little bit variation among different formulations due to changes in the proportion of tablettose as it was replaced with other excipients having low flow ability. Good flow characteristics have been confirmed by flow of granules from hopper during compression and a narrow range of weight variation in compressed tablets. The whole powder blend for all formulations was free flowing (angle of repose less than 32°, Hausner ratio below 1.15 and Carr’s index less than 12.01). During compression flow of the granules was uniform and a very low weight variation was observed in compressed tablets. Lubrication of the granules was very good as the surface of the tablets from all batches was smooth and shiny.

Table 9.

Evaluation of powder blend.

Property	ED-01	ED-02	ED-03	ED-04	ED-05	ED-06	ED-07	ED-08	ED-09	ED-10	ED-11	ED-12
Volume bulk	30 ± 0.87	30 ± 0.79	30 ± 1.0	30 ± 1.0	30 ± 0.76	30 ± 0.57	30 ± 0.88	30 ± 0.60	30 ± 0.80	30 ± 0.78	30 ± 0.5	30 ± 0.8
Volume tapped	27.75 ± 0.46	27.5 ± 0.5	27 ± 0.42	27.4 ± 0.69	27.1 ± 0.7	27 ± 1.0	27.8 ± 0.4	27.15 ± 0.77	26.8 ± 0.45	27.2 ± 0.84	27.6 ± 0.63	26.8 ± 0.49
Bulk density	0.833 ± 0.02	0.833 ± 0.03	0.832 ± 0.02	0.833 ± 0.04	0.826 ± 0.08	0.83 ± 0.06	0.833 ± 0.04	0.833 ± 0.08	0.838 ± 0.05	0.836 ± 0.04	0.836 ± 0.07	0.833 ± 0.08
Tapped density	0.901 ± 0.05	0.909 ± 0.03	0.924 ± 0.07	0.912 ± 0.05	0.915 ± 0.03	0.922 ± 0.02	0.899 ± 0.03	0.921 ± 0.01	0.933 ± 0.02	0.923 ± 0.06	0.909 ± 0.01	0.933 ± 0.04
Hausner ratio	1.082	1.091	1.11	1.095	1.078	1.111	1.079	1.106	1.113	1.104	1.087	1.104
Carr’s index	8.163	9.213	11.058	9.484	10.775	11.084	7.923	10.564	11.336	10.407	8.732	12.005
θ Repose	23.802 ± 0.5	26.41 ± 0.28	28.96 ± 0.07	27.12 ± 0.13	29.46 ± 0.1	28.59 ± 0.2	21.94 ± 0.18	28.73 ± 0.2	29.81 ± 0.2	28.77 ± 0.1	23.63 ± 0.17	30.21 ± 0.3

Results presented as average ± standard deviation (n = 3).

θ Repose, angle of repose.

Hausner ratio and Carr’ index were calculated from average bulk density and average tapped density of each formulation.

3.6. Tablet evaluation

Physically tablets from all the batches were very elegant. Their surface was smooth and shiny without sticking and picking indicating proper lubrication of powder blend. Theoretically weight per tablet was 600 mg with allowed official variation ±5% (British Pharmacopoeias, 2008). Weight variation was very low i.e. less than ±3.5% for all the formulations. The highest weight variation was observed in the case of ED-12 which was ±3.4%. It is well within the official range. Rest of all the formulations showed a weight variation less than ±3.4%. It indicates that the flow of the granules was very efficient and uniform.

Thickness of the tablets from all formulations was within the range of 3.5–3.8 mm. No significant variation was observed in the thickness of tablets from different formulation.

Friability of the tablets from all the formulations was within the official limits (British Pharmacopoeias, 2008) that was less than 0.8%. It was the highest for ED-02 and ED-08 and was 0.45% while for the rest of the formulations it was even below 0.45% (in the range of 0.15–0.3%). No edging was observed in tablets from any formulation. Low friability showed good mechanical resistance of the tablets.

Crushing strengths of the tablets were in the range of 6–10 kg. It is evident from Table 2 that the hardness of the tablet increases with increase in the percentage of tablets. It was the highest for ED-09 which was 9.354 (n = 10) containing 56.83% of tablettose. Its tablettose content was less than ED-01, ED-02, ED-08, and ED-10. Even then its hardness was more than the rest of the formulation and its friability was just 0.3%. As a whole tablets from all the formulations were hard enough to withstand handling during processing.

The drug content of all the formulations was within the range of 97–102%, as in Table 10, which was within the official limits i.e. 95–105% (British Pharmacopoeias, 2008). These results of the drug content showed that the drug has been uniformly blended with excipients.

Table 10.

Tablets evaluation.

Property	ED-01	ED-02	ED-03	ED-04	ED-05	ED-06	ED-07	ED-08	ED-09	ED-10	ED-11	ED-12
Crushing strength	9.201 ± 1.5	8.61 ± 1.8	6.675 ± 1.7	7.055 ± 1.4	6.543 ± 0.9	8.64 ± 1.2	9.15 ± 1.8	8.931 ± 1.67	9.354 ± 1.38	8.422 ± 1.7	7.13 ± 1.3	7.25 ± 1.83
Tensile strength	0.123	0.113	0.087	0.093	0.087	0.117	0.124	0.121	0.127	0.114	0.098	0.099
Friability	0.3 ± 0.15	0.45 ± 0.15	0.15 ± 0.1	0.15 ± 0.15	0.15 ± 0.1	0.3 ± 0.15	0.3 ± 0.15	0.45 ± 0.1	0.3 ± 0.15	0.3 ± 0.15	0.3 ± 0.15	0.3 ± 0.15
Effervescence time	87 ± 3	71 ± 4	54 ± 2	63 ± 5	32 ± 5	43 ± 3	75 ± 5	63 ± 5	52 ± 4	49 ± 4	29 ± 5	41 ± 3
Weight variation	±02	±02	±03	±02.5	±03	±03	±02.5	±02.6	±02.8	±02.5	±02	±3.4
Moisture content	1.79 ± 0.2	1.57 ± 0.5	1.49 ± 0.4	1.36 ± 0.08	1.42 ± 0.07	1.77 ± 0.09	1.83 ± 0.06	1.61 ± 0.1	1.73 ± 0.09	1.47 ± 0.2	1.3 ± 0.26	1.8 ± 0.1
Wetting time	181 ± 3	170 ± 3	166 ± 4	150 ± 3	146 ± 2	150 ± 4	192 ± 3	176 ± 2	184 ± 3	168 ± 2	150 ± 2	159 ± 3
Drug content	97.35 ± 0.93	101.19 ± 0.37	99.72 ± 1.03	100.53 ± 0.87	98.11 ± 0.96	97.26 ± 0.8	99.67 ± 1.06	98.42 ± 1.1	97.9 ± 0.78	99.32 ± 0.99	101.76 ± 0.89	98.79 ± 1.17

Results are presented as average ± standard deviation.

The surface of the tablets was not prone to premature effervescence. As percentage of acid–base pair in tablets was reasonable (20% w/w) they have been uniformly mixed with the rest of the excipients after pulverization into a fine particle size. Reduction in particle size caused increased surface area which leads to a better and quicker reaction between acid and base when exposed to water.

The moisture content of all formulations was within the range of 1.3–1.8%. It was not high enough to cause any premature effervescence. A higher moisture content was observed for ED-12 which was 1.8%.

Effervescence time of all formulations has been presented in Fig. 4. Effervescence time of the tablet containing only citric acid and sodium bicarbonate (ED-01) was 78 s (n = 6). Effervescence reaction was very slow and gradual while the effervescence time of the tablet containing tartaric acid and sodium bicarbonate, in the same concentration, alone without any super disintegrant, was 52 s (ED-07). It proved that reaction between citric acid and sodium bicarbonate was slower as compared to reaction between tartaric acid and sodium bicarbonate.

Figure 4.

Effervescence time of tablets.

3.7. Effect of tablet surface area on disintegration time of tablet

A decrease in surface area available for effervescence reaction resulted in a large increase in effervescence time of all formulations, irrespective of acid base pair and super disintegrant added to the formulation as presented in Table 11. An important factor which can affect effervescence time is compression of the tablet. This factor was nullified by compressing two sized tablets under similar tensile strength and specific hardness (De Jong, 1987). Hardness of the small sized tablet was kept in the range of 5–7.5 kg as evident from Table 12. At this hardness their tensile strength and specific hardness were almost same as those of the larger sized tablets (f₂ = 99.5 for all formulations). An increase in effervescence time due to a larger surface area was in the range of 192.96–307%, as given in the Table 11. The highest increase was observed with ED-11 which was 306.25% of the effervescence time of tablets with a large surface area. The smallest was 192.96% with ED-03. The rest of the increase in disintegration time was in between these two values. It is evident from Table 11, that decreasing the tablet size caused a huge increase in effervescence time of tablets. A comparison of disintegration time of both sized tablets has been graphically presented in Fig. 5.

Table 11.

Effect of tablet surface area on effervescence time of the tablet.

Batch No.	Effervescence Time (s)		% Increase in effervescence time
	Large punch	Small punch
ED-01	78	163	2008.97
ED-02	54	126	233.33
ED-03	71	137	192.96
ED-04	63	142	225.4
ED-05	32	98	306.25
ED-06	53	112	211.32
ED-07	52	119	228.85
ED-08	44	97	220.45
ED-09	75	153	204
ED-10	49	112	228.57
ED-11	29	87	300
ED-12	41	102	248.78

E. Time, effervescence time.

Table 12.

Comparison between mechanical properties of tablets compressed on larger (13 mm) and smaller (10 mm) punch.

Batch No.	Larger punch size (13 mm)					Smaller punch size (10 mm)
	K (kg)	T (mm)	D (mm)	T.S.	τ	K (kg)	T (mm)	D (mm)	τ	T.S.
ED-01	9.201	3.65	13	0.123	0.194	7.14	3.5	10	0.204	0.13
ED-02	8.61	3.72	13	0.113	0.178	6.67	3.45	10	0.193	0.123
ED-03	6.675	3.75	13	0.087	0.137	5.35	3.45	10	0.155	0.099
ED-04	7.055	3.7	13	0.093	0.147	5.78	3.6	10	0.161	0.102
ED-05	6.543	3.65	13	0.087	0.138	5.52	3.55	10	0.155	0.099
ED-06	8.64	3.6	13	0.117	0.185	6.99	3.5	10	0.2	0.127
ED-07	9.15	3.61	13	0.124	0.195	7.47	3.6	10	0.208	0.132
ED-08	8.931	3.6	13	0.121	0.191	7.58	3.5	10	0.217	0.138
ED-09	9.354	3.6	13	0.127	0.199	7.5	3.55	10	0.211	0.134
ED-10	8.422	3.6	13	0.114	0.18	6.65	3.46	10	0.192	0.122
ED-11	7.132	3.55	13	0.098	0.154	5.99	3.5	10	0.171	0.109
ED-12	7.249	3.58	13	0.099	0.156	5.9	3.48	10	0.17	0.108

K, compression force; T, tablet thickness; D, tablet diameter; T.S., tensile strength of tablet; τ, specific hardness of tablets.

Figure 5.

Effect of surface area on effervescence time.

3.8. Effect of disintegrant on effervescence time of tablets

When disintegrant was added along with an acid/base pair, it enhanced effervescence reaction. It was much more vigorous as compared to that without disintegrant. Tablet moved up and down during the whole of the effervescent reaction. Cross carmellose sodium acted as a wicking agent increasing water penetration into the inner core of the tablets (Handbook of Pharmaceutical Excipients and Edition, 2009). The acid/base pair got exposed to water quickly and the rate of the effervescent reaction was enhanced. In a higher concentration cross carmellose sodium absorbed water and formed a gel like material. The core of the tablet remained intact and the inner portion of the tablet got slowly exposed to water and the rate of effervescence reaction got reduced. It indicated that cross carmellose sodium was efficient at low concentration (3%, w/w) as compared to the high percentage (5%, w/w).

SSG produced a concentration dependent decrease in effervescence time with both CA/SBC and TA/SBC pairs. At lower concentrations (3% w/w) a drop in disintegration time by SSG was smaller than that caused by the same concentration of cross carmellose sodium. But at a higher concentration (5% w/w), SSG was more efficient than cross carmellose sodium. A drop in disintegration time by SSG at a higher concentration was larger than the drop caused by cross carmellose sodium with both acid/base pairs.

3.9. Effect of tablet compression force

Compression force has a negligible effect on the effervescence time of the tablet as evident from the data presented in Table 13. When the tablet breaks down during effervescence reaction, the surface area available for effervescence reaction increases resulting in its enhanced rate. By increasing the crushing strength of the tablet, water penetration into the tablet reduces but as the effervescence reaction starts, it moves into the tablet’s core layer by layer, overcoming the hard tablet core. As effervescent tablets contain very low moisture they are very much prone to capping and edging at a higher level of compression force.

Table 13.

Effect of tablet hardness on tablet effervescent reaction.

Parameter (Unit)	Level-1	Level-2	Level-3
Average hardness (kg)	5.68 ± 0.49	9.21 ± 1.27	12.34 ± 0.85
Weight (mg)	607.52 ± 1.36a	603.86 ± 1.4a	604.37 ± 1.51a
Thickness (mm)	3.93 ± 0.074	3.70 ± 0.059	3.61 ± 0.063
Effervescence time (s)	38 ± 4	45 ± 3	52 ± 3

Results are presented as average ± standard deviation.

Level-1, 4–7 kg; Level-2, 7–12 kg; Level-3, 12–14 kg.

^aWeight variation (%).

4. Conclusion

It is concluded from the study that the SeDeM expert system can be successfully applied for the prediction of suitability of material for direct compression. It gives accurate predictions about material behavior and response of the material was same as predicted by the SeDeM expert system. It provides information about shortcoming of the material to be processed by direct compression which can be rectified at a pre formulation level to get a robust formulation that can be easily scaled up for commercial manufacturing. The SeDeM expert system also reduces the number of trials at a pre formulation level to get produced by direct compression especially in the case of a high drug load. By developing a database of the excipients commonly used in pharmaceutical formulation, the material of the desired characteristics can be selected with particular characteristics.

Effervescent tablets are highly moisture sensitive and even a trace amount of water can result in complete deterioration of the product. Direct compression is the most preferable method for the preparation of effervescent tablets. By applying the SeDeM expert system in the formulation of effervescent tablets, commercial manufacturing of the dosage form will become very economical and time saving.