Design Principles for Hot Air Distributors in Spray Drying

Due to its advantages—including a simple process, the ability to handle heat-sensitive materials, and scalability—spray drying has been widely adopted in many fields. Since the reform and opening-up, a large number of specialized drying equipment manufacturers have emerged in China. Over the past decade, spray drying technology has made significant progress, with product quality now comparable to that of world-renowned manufacturers. This technology not only meets the needs of China’s light chemical and environmental protection industries but has also expanded into overseas markets.      

For a long time, attention to spray drying systems has generally focused on:      

⑴ Selection of atomizers (units);      

⑵ the provision of sufficient airflow and heat;      

⑶ powder recovery and discharge.      

Wang Xizhong et al. pointed out: “A successful spray dryer design should include hot air inlet and outlet configurations compatible with the atomizer, as well as a hot air distribution system” [1]. K. Master’s also noted that the rate of moisture evaporation within the drying tower increases as the relative velocity between the droplets and the hot air increases [2].

Tang Jinxin et al. proposed three important principles regarding the design requirements for hot-air distributors [3], all of which emphasize the importance of hot-air distribution in spray drying.

In subsequent installations, it was found that most companies still did not attach sufficient importance to this issue, merely achieving a “similar appearance” in terms of structure without truly grasping the essence, resulting in the following situations:      

(1) Significant temperature differences within the same cross-section of the tower, leading to localized material adhesion to the tower walls; (2) Due to improper gas-liquid phase contact, the drying intensity decreased significantly, resulting in the drying tower becoming increasingly larger;     

⑶ In a drying tower with a processing capacity significantly reduced from the original design, failure to consider the flow velocity range of the hot air distribution lowered the drying intensity, and the material continued to adhere to the tower walls in large quantities; ⑷ Thermal efficiency was very low, making it difficult to reduce the outlet air temperature.      

Therefore, we believe that the correct design of the hot air distributor directly affects the success or failure of the drying system’s operation. Building on previous knowledge, this paper proposes a rational method for gas-liquid phase contact, with the aim of providing a correct analysis and guidance for the design of hot air distributors.

1 Theoretical Basis

K. Masters [2] proposed the following equations describing the evaporation of droplets under relative velocity conditions:      

Mass transfer: Sh = 2 + K₁R_(ex)S_(cy) (1)      

Heat transfer: Nu = 2 + K₂ReX’Pry’ (2)      

Where: the Sherwood number Sh = KgD/Dv, the Nusselt number Nu = hcD/Kd, the Schmidt number Sc = μa/Dvρa, the Prandtl number Pr = Cpμa/Kd, and the Reynolds number Re = Dvρa/μa. D is the droplet diameter, ρa is the density of the dry medium, μa is the viscosity, Cp is the specific heat capacity at constant pressure, Kd is the average thermal conductivity of the gas film surrounding the droplet, hc is the convective heat transfer coefficient, Kg is the mass transfer coefficient, and Dv is the diffusion coefficient.

The values of x, y, x’, y’, K₁, and K₂ in Equations (1) and (2) are still subject to debate; most researchers tend to adopt:      

x = x’ = 0.5 (3)      y = y’ = 0.33 (4)      

The value of x in Equation (3) is an average value that increases as Re increases; when Re increases from 1 to 104, x increases from 0.4 to 0.6. Unfortunately, the experimental ranges for Equations (1) through (4) all have Re values not exceeding 1000. However, it is already evident that the mass transfer and heat transfer coefficients for drying increase with increasing Re; that is, assuming the properties of the drying medium and the material being dried remain constant, Re plays a significant role. The relative velocity v can be considered to have a direct influence on Re. In traditional calculations of convective heat transfer coefficients for liquids without phase change, the Dittus and Boelter correlations [4] are commonly used:      

Nu = 0.023Re⁰.⁸Pr⁰.⁴ (5) or (6) where α is the heat transfer coefficient; λ is the thermal conductivity of the liquid; d is the particle diameter; v is the relative gas-liquid flow velocity; μ is the dynamic viscosity of the liquid; Cp is the specific heat capacity at constant pressure; and ρ is the density of the liquid.

In this equation, Re ≥ 10,000 and 0.7 < Pr < 120. A comparison of Equation (1) with Equation (5) shows that the exponent can be increased from 0.4 to 0.8 within the turbulent flow range of the Re number. This explains the observation emphasized by K. Master et al. that “the rate of water evaporation increases with the relative velocity between the droplets and the air.” When the Re number is within the turbulent range, the relationship is approximately proportional to the 0.8th power.      

2. Performance Comparison of Common Hot-Air Distributors Among the types of hot-air distributors used in spray drying, the following designs have been employed:

(1) Uniform downward flow distributed from the ceiling at the top of the tower      

This design assumes that as long as the air is introduced uniformly and there is sufficient heat, the drying objective can be achieved. The empty tower velocity in the drying tower is only 0.5–0.8 m/s; even when the tower top is narrowed, the outlet air velocity remains at only 10 m/s, generally maintaining a laminar flow state. There is no direct contact between the hot air and the atomized droplets. This configuration is found not only in domestic systems but also in many imported units. The result is an oversized tower and reduced efficiency.      

(2) To prevent wall adhesion, the hot air is divided into two or three streams      

Designers believe that as long as hot air flows along the tower wall, it can prevent un-dried droplets from colliding with the wall and causing wall adhesion. In reality, the hot air velocity at the edge cannot be very high, and the velocity of the droplets when they reach the tower wall will not be very high either. Consequently, the relative velocity between these two fluid streams is very low, making rapid drying difficult to achieve, and wall adhesion will still occur. The hot air along the tower wall is either ineffective or has little effect.      

The well-known MD-type tower employs cold air blowing, which is beneficial for ensuring material quality. In reality, by this point, the droplets have already completed the “constant-rate” drying phase (at least the particle surfaces have dried), which has no direct connection to wall adhesion [5].      

Of course, the form of wall adhesion is also related to the atomizer’s spray distance, the design of the drying tower, and the material’s glass transition temperature, among other factors. These issues have been discussed in detail in [1]. Distributing the hot air reduces the hot air volume in the central zone, thereby lowering the flow velocity and resulting in reduced hot air utilization efficiency.      

(3) Mismatch Between the Hot Air Distributor and the Atomizer      

Using a rotating airflow with a nozzle-type atomizer, or a straight-through airflow with a rotary atomizer. Both of these configurations have been observed in production; the result can only be wall sticking or a significant drop in thermal efficiency, which is clearly incorrect.      

3. The Importance of Central Hot Air at the Tower Top      

When any atomizer is operating, the flow velocity of the droplets is highest immediately upon leaving the atomizer outlet. As the droplets travel through the air, their velocity rapidly decays due to air resistance; the initial velocity can reach 130 m/s, while the final velocity can approach zero, This requires us to accurately determine, based on Equations (1) through (4), where the hot air should come into contact with the droplets in order to achieve optimal mass and heat transfer rates.      

Since (most) atomizers are designed to be located at the center of the tower top, the hot air should be concentrated in the center so that an airflow—equivalent to a turbulent flow—rapidly impacts the droplet cluster; the air volume and heat input depend on the amount required to dry the moisture on the particle surfaces. The remainder can be uniformly distributed within the tower to complete the drying process in the subsequent deceleration sections. As long as the moisture on the particle surfaces can be dried quickly, wall sticking in the tower can be largely prevented.      

The closer the high-velocity airflow is to the atomizer nozzle, the higher the drying efficiency. However, when considering airflow velocity, one must also take into account the relationship between pressure drop and the square of the velocity; higher air velocity is not necessarily better. Furthermore, higher air velocities cause the droplet clusters to drift downward, resulting in the loss of some effective drying space. Specific parameters depend on the characteristics of various materials. However, the general trend is to utilize the high-velocity zone where the gas and liquid phases meet to rapidly dry the droplet surfaces, thereby achieving the evaporation of most of the moisture—this is how the true advantages of spray drying are realized.      

4. Key Elements of an Effective Hot Air Distributor      

(1) To ensure good contact and mixing between the gas and liquid phases, the gas must first be distributed uniformly. To achieve uniform distribution, two methods have been previously described:

① In the design of a rotary atomizer system, a logarithmic spiral volute [3] must be used to ensure that the hot air entering the volute on one side is evenly distributed into the tower via the volute and its internal baffles.

② For hot air distribution in direct-flow atomizers, various direct-flow baffles [1] can be employed, but a direct-flow nozzle atomizer must be used.

⑵ The outlet of the hot air distributor should be positioned as close as possible to the atomizer’s liquid spray outlet, with the angle between the two directions approaching 90°, to increase shear force. The advantages of the turbulent flow phase should be utilized to shorten the drying time.      

⑶ When the flow velocity at the hot air distributor outlet is too high, resistance increases in proportion to the square of the velocity; therefore, “pressure drop within the system” must be taken into account, and the air velocity should be selected with care.  

5. Conclusion      

In recent years, a trend toward indiscriminately increasing the volume of spray drying units has been observed in their design and manufacturing. This not only negates the advantage of short spray drying times but also increases construction costs and the floor space (or volume) occupied by the equipment, which is disadvantageous to users.      

When the hot air distributor and atomizer are properly configured, the volume of the drying tower should fall within a reasonable range without significant variation. Larger is not necessarily better. With advances in technology and the application of various intensification measures, drying towers will inevitably become smaller.      

The hot air distributor is an important aspect, but it is not the only factor. Therefore, in the design of spray dryers, selection should be based on the characteristics of various materials and a comprehensive evaluation of all parameters to achieve optimal system performance.