Why Light Hydrocarbon Distillation Towers Always Carry Water at the Top

2025-12-26

1. A Common Problem in Almost All Light Hydrocarbon Units

In light hydrocarbon distillation units, engineers frequently encounter the following phenomenon:

Even when the feed contains only trace amounts of water, it can travel with the hydrocarbon vapor all the way to the top of the column.

This seems counterintuitive. Water boils at 100 °C at atmospheric pressure, whereas propane and butane boil at −42 °C and −0.5 °C, respectively.

Intuitively, one would expect that with enough trays and sufficient reflux, the water would be “washed” back to the column bottom.

In practice, however, water still appears at the top, and paradoxically, increasing the reflux ratio often makes water carryover worse.

Interestingly, in methanol–water distillation, as long as separation is sufficient, the top product can be almost completely water-free.

Why is water nearly impossible to block in light hydrocarbon systems, while it can be entirely removed in methanol systems?


2. Let’s Be Clear: This Is Not a Separation Efficiency Issue

Before discussing the mechanism, an important engineering conclusion must be emphasized:

Water carryover at the top of light hydrocarbon columns is not necessarily due to operational or design issues, but is often determined by phase equilibrium limits.

If the following conditions are met:

  • Hydrocarbons and water are immiscible
  • Vapor–liquid equilibrium is established in the column

Then the maximum water fraction the top can carry can be calculated directly at the given top temperature and pressure. Increasing the reflux ratio cannot surpass this physical limit.


3. Key Mechanism: “Simultaneous Vaporization” in Immiscible Systems

The fundamental difference between light hydrocarbon–water and methanol–water systems lies in miscibility.

For systems where water and hydrocarbons form separate immiscible layers:

  • At a given temperature, the total system vapor pressure equals the sum of the saturated vapor pressures of each liquid phase.
  • When the total vapor pressure equals the operating pressure, both phases vaporize simultaneously.
  • The proportion of each component entering the vapor phase is determined by its own saturated vapor pressure, not by its relative boiling point.

4. An Intuitive Engineering Example: Butane–Water System

Consider a butane–water system at 210.3 kPa and 20 °C:

  • Butane’s saturated vapor pressure ≈ 208 kPa
  • Water’s saturated vapor pressure ≈ 2.3 kPa

Total pressure: 208 + 2.3 ≈ 210.3 kPa

Boiling occurs, and because butane and water form separate layers, both vaporize simultaneously in the ratio ≈ 208:2.3.

This means that even though water’s boiling point is far higher than butane’s, it still inevitably enters the ascending vapor as a small fraction, fully dictated by phase equilibrium—not tray number or reflux ratio.


5. How Much Water Can the Top Actually Carry?

Example: Propane–Butane Column

  • Top pressure: 1.5 MPa (absolute)
  • Top temperature: 44 °C
  • Saturated vapor pressure of water at 44 °C: 9.1 kPa

Since hydrocarbons and water are immiscible, and free water exists at the top, the partial pressure of water in the vapor can be approximated by its saturation pressure.

Maximum mole fraction of water at the top:

$$ y_{\mathrm{H_2O}}^{\max} \approx \frac{P_{\mathrm{H_2O}}^{\mathrm{sat}}}{P_{\mathrm{top}}} = \frac{9.1}{1500} \approx 0.0061 $$

≈ 0.61 mol% water. This is the physical limit determined by temperature and pressure.


6. Engineering Formula for Maximum Water Carryover

Let:

  • $G_D$ = molar flow of top product
  • $G_R$ = molar flow of top reflux
  • $G_{\text{top}} = G_D + G_R$ = total vapor flow entering the condenser

Then the maximum water flow carried by the top vapor is:

$$ G_{\mathrm{H_2O}}^{\max} = \frac{P_{\mathrm{H_2O}}^{\mathrm{sat}}}{P_{\mathrm{top}}} (G_D + G_R) $$

Substituting the numbers:

$$ G_{\mathrm{H_2O}}^{\max} \approx \frac{9.1}{1500} (G_D + G_R) $$

This represents the top’s maximum theoretical water-carrying capacity under these operating conditions.


7. Three Engineering Conclusions

1️⃣ When is the bottom guaranteed to be water-free?

If the feed water content is less than the top’s maximum water-carrying capacity and the condenser/reflux drum can completely remove free water, then all feed water will be carried away by the top vapor.
The bottom product remains water-free.

2️⃣ Why does increasing the reflux ratio make top water carryover worse?

Because:

$$ G_{\text{top}} = G_D + G_R $$

Increasing reflux → higher total top vapor flow → higher theoretical water-carrying capacity.
Phase equilibrium does not change; only the vapor pathway for water is amplified.

3️⃣ Water content on trays along the column

  • Vapor moves up the column, and temperature decreases along the way.
  • Water’s saturation pressure drops, reducing the vapor’s water-carrying ability.
  • With small feed water, condensation may occur at some intermediate tray and drain back to the bottom.

8. Engineering Measures: How to Ensure Water-Free Bottom Product

To make the bottom completely water-free, the core goal is to ensure all feed water is carried by the top vapor and removed in the condenser.

Key measures include:

1. Control feed water content

  • Feed water must be less than the top’s theoretical maximum water-carrying capacity ($G_{\mathrm{H_2O}}^{\max}$).
  • Otherwise, water will appear at the bottom even with perfect cut-water measures.

2. Top liquid–liquid separation and water cut

  • Equip the top reflux drum with reliable liquid–liquid separation.
  • Remove condensed water completely, preventing it from returning into the column.
  • This is the most critical step to ensure all feed water is carried away.

3. Tray water cut points

  • Install water cut points at appropriate trays to remove condensed water.
  • Prevent local water accumulation or entrainment along the column.

4. Reflux ratio effect on top water-carrying capacity

  • Increasing reflux increases total top vapor flow, enhancing the top’s theoretical water-carrying capacity.

To make calculations easier, We developed an app that can calculate the saturated vapor pressure for hundreds of chemical substances. You can download it here: Vapor Pressure Calculator App


← Previous: Raoult's Law vs Henry's Law — Reducing Material Loss from NCG Emissions

Author: Hadel
Published: 2025-12-26
Source: chem.zhanghd.fun