Intermittent drying: Can it reduce energy consumption by 50% while maintaining moisture control?

In a continuous coffee dryer, hot air is supplied evenly throughout the drying process. This is a familiar and easily controlled operation, but there is a lesser-known issue: coffee beans do not necessarily need continuous heat for the moisture-removal process to be effective.

A study published in June 2026 in the Journal of Food Process Engineering compared continuous drying with intermittent drying on Arabica. The notable result was that the intermittent drying regime at 55°C, combined with a 5-minute break, reduced energy consumption by approximately 50% compared to the corresponding continuous drying condition in the experiment.

What happens during the time the machine stops supplying heat? And why can a short break affect the drying efficiency?

Coffee beans do not dry evenly from the outside in

When coffee is fed into the dryer, heat from the air is transferred to the surface of the beans. Water near the surface can evaporate first, while water deeper within the bean structure takes time to move out.

The drying process therefore involves two related phenomena:

• Heat transfer: energy from hot air is transferred to the coffee beans.
• Moisture transfer: water moves from within the bean to the surface, where it evaporates into the environment.

If only looking at air temperature, the operator might think that increasing heat will always result in faster drying. However, moisture removal speed also depends on the ability of water to diffuse within the bean, ambient humidity, air velocity, bean size, and material state.

In the later stages of the drying process, moisture removal speed often slows down because water must move from deep within the material to the surface. This is one reason why continuous heat supply may not always result in optimal energy efficiency.

How does intermittent drying differ from continuous drying?

Continuous drying maintains heat supply throughout the cycle. Intermittent drying divides the process into heat supply periods and rest or temperature-humidity adjustment periods.

During the rest period, the device does not continue to supply heat according to the same regime. However, this does not mean that the entire moisture transfer process comes to a complete stop. Water within the bean can still redistribute, reducing the moisture gradient between regions.

This rest period is often called tempering. It is the stage that allows moisture within the material to redistribute before continuing the drying process.

If designed properly, intermittent drying can improve moisture uniformity and prevent surface over-drying while the interior remains moist.

What conditions did the 2026 Arabica study experiment with?

Oliveira and colleagues published the study Modeling the Drying Kinetics of Arabica Coffee: A Comparative Study of Continuous and Intermittent Drying Regimes on 08/06/2026.

The research group conducted experiments at three temperature levels:

The data on moisture content over time were analyzed using three groups of models: first-order models, Page models, and fractional derivative models.

The results show that temperature has a significant impact on the drying kinetics. Intermittent drying produces a smoother moisture reduction curve and supports the redistribution of moisture during the tempering phase.

Why was the fractional derivative model used?

In the classical drying model, the diffusion process is typically described based on Fick's laws. These models can be useful, but sometimes they do not fully describe the complex moisture transport phenomena in heterogeneous biological materials.

The fractional derivative model adds the ability to describe processes with historical dependence. In other words, the current rate of state change can be influenced by what has occurred previously.

In this study, the fractional derivative models reduced the mean squared error by approximately 40–70% compared to the classical first-order model, depending on the drying conditions.

The estimated fractional order parameter ranged from 0.80 to 1.46. This result indicates that the moisture transport process in the experimental system is not always well described by a simple classical diffusion model.

This is a modeling tool to understand and predict the drying curve, not evidence that all coffee dryers need to use the fractional derivative algorithm in their control systems.

What does the 50% energy reduction mean?

The most notable result is that the intermittent drying regime at 55°C with a 5-minute tempering time helps reduce energy consumption by approximately 50% compared to the corresponding continuous drying regime under the experimental conditions.

The key point lies in the phrase under the experimental conditions corresponding. This is not a commitment that all processing facilities will reduce their electricity or fuel bills by 50% when switching to intermittent drying.

What else affects the actual effectiveness?

• The initial moisture content and the amount of coffee fed into the dryer.
• The type of dryer, its insulation capabilities, and airflow distribution.
• The relative humidity of the incoming air.
• The frequency and duration of heat supply and rest periods.
• The energy costs, productivity, and completion time of a batch.

If the rest period is too long, the total processing time may increase. If the temperature is too high, shortening the drying time may have to be traded off against the risk of affecting the quality. Therefore, it is necessary to evaluate the energy consumption, time, moisture uniformity, and coffee quality simultaneously.

Should the 55°C and 5-minute rest regime be applied immediately to Vietnamese coffee?

It is not advisable to take the parameters of this study as a fixed formula for all types of raw materials.

Research used Arabica and specific experimental conditions. Vietnamese coffee may differ in terms of variety, processing method, initial moisture, bean size, residual tacky layer, and equipment conditions.

To conduct experiments at a processing facility, a controlled comparative process can be established:

1. Divide the same batch of raw materials into groups of equivalent weight.
2. Maintain the original input conditions and equipment type, only changing the drying mode.
3. Record temperature, air humidity, heat-up time, and rest time.
4. Measure moisture at multiple locations within the batch, rather than just taking a surface sample.
5. Compare total energy consumption per kilogram of coffee to reach the target moisture level.
6. Roast and evaluate sensory samples after drying to determine the impact on quality.

In practice, a drying mode is only valuable when it achieves the target moisture level, maintains batch uniformity, and does not compromise sensory quality.

Efficient drying is not just about quickly evaporating water.

The coffee drying process is a balancing act between heat transfer rate, moisture diffusion, energy consumption, and quality preservation. Continuous drying has the advantage of simple operation, but intermittent drying offers the possibility of optimizing heat-up time and allowing moisture to redistribute within the beans.

A 2026 study showed that a well-designed rest period can improve energy efficiency in specific conditions. This is an area of interest for processing facilities that want to reduce operating costs while maintaining good coffee quality.

References

1. Oliveira, D. M. et al. (2026). Modeling the Drying Kinetics of Arabica Coffee: A Comparative Study of Continuous and Intermittent Drying Regimes. Journal of Food Process Engineering. First published 08 June 2026. https://doi.org/10.1111/jfpe.70625
2. Duque-Dussán, E. (2026). Development and Validation of the D²MCD Family: Dual-Term Diffusion Models for Coffee Drying. Journal of Food Process Engineering. https://doi.org/10.1111/jfpe.70459
3. Romero Leiton, J. P. et al. (2025). Mathematical modeling of coffee beans drying: addressing challenges of determining moisture ratio under natural conditions. Coffee Science, 20, e202385. https://doi.org/10.25186/.v20i.2385