Enhancement of conventional distillation configurations for ternary mixtures separation


Rigorous simulations at steady-state are conducted for the separation of two different ternary mixtures by conventional distillation with direct and indirect separation sequence. The studied chemical systems are (ethanol/n-propanol/n-butanol) and (benzene/toluene/m-xylene) with 99 mol% products purity at different feed compositions: (45/10/45), (33.3/33.3/33.3), and (10/80/10). In order to reduce the remixing effects in the first column a sidedraw stream is introduced at peak point composition of middle component as transfer stream to the second column. Economic optimization is carried out for the conventional and improved configurations based on energy consumption and total annual cost (TAC) as the objective function. The results indicate that improvements of thermodynamic performance of both columns are achievable and the maximum TAC saving is 29% in case of indirect sequence and 19% for direct sequence at low concentration of middle component. Improved conventional distillation sequences show a significant reduction of flue gas emissions (CO2, SOx, and NOx) whish is generated by the combustion process of the utility systems.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4


  1. Agrawal R, Fidkowski Z (1998a) Are thermally coupled distillation columns always thermodynamically more efficient for ternary separations? Ind Eng Chem Res 37:3444–3454

    Article  CAS  Google Scholar 

  2. Agrawal R, Fidkowski Z (1998b) Improved direct and indirect systems of columns for ternary distillation. AIChE J 44:823

    Article  CAS  Google Scholar 

  3. Biegler LT, Grossmann IE, Westerberg AW (eds) (1997) Systematic methods for chemical process design. Prentice-Hall, NJ

    Google Scholar 

  4. Douglas JM (ed) (1988) Conceptual design of chemical process. McGraw-Hill Inc, New York

    Google Scholar 

  5. Doherty MF, Malone MFJ (eds) (2001) Conceptual design of distillation systems. McGraw-Hill Inc, New York

    Google Scholar 

  6. Emtir M, Essllaa O, Fonyo Z (2005) Optimisation of energy-integrated distillation schemes for BTX-separation. AIDIC Conf Ser 7:115–124

    Google Scholar 

  7. Glinos K, Malone MF (1988) Optimality regions for complex column alternatives in distillation systems. Chem Eng Res Des 66:229

    CAS  Google Scholar 

  8. Halvorsen IJ (1999) Optimal operation of petlyuk distillation: steady-state behaviour. J Process Control 9:407–424

    Article  CAS  Google Scholar 

  9. Hernandez S, Jimenez A (1999) Controllability analysis of thermally coupled distillation systems. Ind Eng Chem Res 38:3957

    Article  CAS  Google Scholar 

  10. Salvador H, Segovia-Hernandez JG, Rico-Ramirez V (2006) Thermodynamically equivalent distillation schemes to the Petlyuk column for ternary mixtures. Energy J 31:2176–2183

    Article  Google Scholar 

  11. Smith R, Delaby O (1991) Targeting flue gas emissions. Trans IChemE 69:492

    CAS  Google Scholar 

  12. Turton R, Bailie RC, Whiting WB, Shaeiwitz, JA (eds) (1998) Analysis, synthesis, and design of chemical process. Prentice-Hall, NJ

Download references


This work is gratefully acknowledged by Libyan Petroleum Institute, group of separation technology.

Author information



Corresponding author

Correspondence to Mansour Emtir.


Appendix A

Total annual cost (TAC) calculations:

Sizing and costing of distillation columns

For a given number of theoretical trays (N), HYSYS simulator calculates column diameter (D) for selected valve tray distillation column with 50.8 mm weir height and valve trays of Glitsch type. In order to estimate the actual number of trays (N actual), overall column efficiency (E 0) is estimated from (Doherty and Malone 2001):

$$ E_{\rm{0}} = 0.24 + 0.76\exp \,((\mu \alpha )^{0.5} ) $$

where μ is the viscosity of the liquid mixture at the feed composition (Centipoises) and α is the volatility between the key components; both are evaluated at the average temperature and pressure in the column. The actual number of trays is estimated by dividing the theoretical number of trays by the overall column efficiency:

$$ N_{\text{actual}} = \frac{N}{{E_{\rm{0}} }} $$

The height of the column (H) for 0.6 m tray spacing, adding 1.2 m at the top for vapor disengagement and 1.8 m at bottom for liquid level, 1.5 m as skirt height and 1.5 m for extra feed space (Biegler et al. 1997) is given by:

$$ H = (N_{\text{actual}} - 1)0.6 + 6 $$

The installed cost of distillation column is estimated applying an installation factor to the purchased cost. The purchase cost consists of shell and tray costs. For carbon steel construction distillation column with valve tray internals, the following cost equations that are updated from mid 1968 to mid 2005 using the ratio of Marshall & Swift index.

$$ {\text{ICCS}} = \left( {\frac{M\& S}{280}} \right)937.61\,D^{1.066} \,H^{0.802} (3.18) $$


installed cost of column shell, \({\$}\).

If the design pressure (P) inside the column is more than 345 kPa, a correction factor of [1 + 0.000145(P-345)] is applied. D and H are the diameter and height of the column in meters.

$$ {\text{ICCT}} = \left( {\frac{M\& S}{280}} \right)97.24\,D^{1.55} h $$


installed costs of column trays, \({\$}\)

where h is the tray stack height (m) for 0.6 m tray spacing:

$$ h = (N_{\text{actual}} - 1) \times 0.6 $$
$$ {\text{Column}}\,{\text{installed}}\,{\text{cost}},\$ = {\text{ICCS}} + {\text{ICCT}} $$

Sizing and costing of heat transfer equipments

The heat transfer area A (m2), of the condenser and reboiler are calculated according to the following Equation:

$$ A = \frac{Q}{{U \times {\text{LMTD}}}} $$

where Q is the heat duty (KJ/h), LMTD is the logarithmic mean temperature difference (°C) and U is the overall heat transfer coefficients (kJ/m2 h °C).

$$ {\text{LMTD}} = \frac{{\Delta T_{2} - \Delta T_{1} }}{{\ln \frac{{\Delta T_{2} }}{{\Delta T_{1} }}}} $$

T 2 is the temperature difference between the inlet streams, and ∆T 1 is the temperature difference between the outlet streams of the heat transfer equipment. Assuming, U = 4320 kJ/m2 h °C for reboilers and 3060 kJ/m2 h °C for condensers. The cost of heat transfer equipment can be correlated as a function of surface area. Assuming shell and tube, floating head, and carbon steel construction, the following cost equation is used:

$$ {\text{ICHE}} = \left( {\frac{M\& S}{280}} \right)474.67\,A^{0.65} (3.29) $$


installed cost of heat exchanger, \({\$}\).

Annual operating and capital costs

The operating costs are assumed to be only utility cost (steam and cooling water costs). The flow rate of cooling water W c, is calculated after the following equation:

$$ Q_{\text{c}} = W_{\text{c}} C_{\text{p}} (T_{\text{co}} - T_{\text{ci}} ) $$

where Q c is the condenser duty (kJ/h), C P is the specific heat of water = 4.181 kJ/kg K, T co and T ci are the outlet and inlet temperatures of cooling water and equals to 318 and 303 K, respectively (Turton et al. 1998).

The flow rate of steam W r, is calculated by the following equation:

$$ Q_{\rm{r}} = W_{\rm{r}} \lambda_{\rm{s}} $$

where Q r is the reboiler duty (kJ/h), λs is the latent heat of steam, equal to 2,083 kJ/kg for low-pressure steam and 2,000 kJ/kg for medium-pressure steam (Table 9).

Table 9 Utility prices for the case study

The capital costs (purchase and installation costs) are annualized over a period referred to as the plant lifetime, and assumed to be 10 years. The operating hours per year is set to be 8,000.

$$ {\text{ACC}} = \frac{{{\text{capital}}\,{\text{costs}}}}{{{\text{plant}}\,{\text{life}}\,{\text{time}}}} $$
$$ {\text{TAC}} = {\text{AOC}} + {\text{ACC}} $$


total annual cost, \({\$}\)/year


annual operating cost, \({\$}\)/year


annual capital cost, \({\$}\)/year.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Emtir, M., Etoumi, A. Enhancement of conventional distillation configurations for ternary mixtures separation. Clean Techn Environ Policy 11, 123–131 (2009). https://doi.org/10.1007/s10098-008-0174-3

Download citation


  • Distillation Column
  • Feed Composition
  • Total Annual Cost
  • Relative Volatility
  • Alcohol System