INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 102

Comparative evaluation of granulometric distribution in grains processed

by ball and hammer Mills.

Análisis comparativo de la distribución granulométrica de granos molidos en molino de bolas y

molino de martillos.

Stefanie Bonilla Bermeo

* ; Fernando Noblecilla Arévalo

; Iván Torres Tapia

; Carlos Valdiviezo Rogel

Received: 10/25/2025 – Accepted: 12/05/2025 – Published: 01/01/2026

Research

Articles

Review

Articles

Essay Articles

* Corresponding

author.

This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 (CC BY-

NC-SA 4.0) international license. Authors retain the rights to their articles and may share, copy, distribute,

perform, and publicly communicate the work, provided that the authorship is acknowledged, not used for

commercial purposes, and the same license is maintained in derivative works.

Abstract.

The grinding of grains is fundamental in industrial processes, where the resulting particle size distribution directly impacts product quality. This study aimed to

compare the granulometric distribution of corn and soybeans processed using a hammer mill, ball mill, and their combination. Samples of corn and soybeans

were ground using three configurations: hammer mill, ball mill, and sequential milling with both. The resulting material was sieved to determine weight retained

per mesh and calculate characteristic diameters (D10, D50, D90). Additional particle microscopy and ANOVA were performed to evaluate significant

differences. The hammer mill produced coarse, heterogeneous distributions, especially for soybeans (D50 ≈ 2.9 mm). The ball mill generated a higher proportion

of fine particles in corn (D50 ≈ 1.38 mm) but was ineffective for soybeans (D50 ≈ 3.53 mm). The mill combination achieved the most uniform distribution for

both grains (D50 ≈ 1.05–1.25 mm). ANOVA detected no global significant differences, though morphological and distributional disparities were observed in

sieve analysis. The combined milling approach optimized granulometric distribution, overcoming the limitations of each individual equipment.

Keywords.

Granulometric distribution, Hammer mill, Ball mill, Corn grinding, Soybean grinding.

Resumen.

La molienda de granos es fundamental en procesos industriales, donde la distribución granulométrica resultante incide directamente en la calidad del producto.

El objetivo fue comparar la distribución granulométrica de maíz y soja procesados en molino de martillo, molino de bolas y su combinación. Se molieron

muestras de maíz y soja utilizando tres configuraciones: molino de martillo, molino de bolas y la secuencia de ambos. El material obtenido se tamizó,

determinándose los porcentajes retenidos por malla y calculándose los diámetros característicos (D10, D50, D90). Adicionalmente, se realizó análisis

microscópico de partículas y ANOVA para evaluar diferencias significativas. El molino de martillo produjo distribuciones gruesas y heterogéneas, especialmente

en soja (D50 ≈ 2.9 mm). El molino de bolas generó un mayor porcentaje de finos en maíz (D50 ≈ 1.38 mm), pero fue ineficaz para soja (D50 ≈ 3.53 mm). La

combinación de molinos logró la distribución más uniforme para ambos granos (D50 ≈ 1.05-1.25 mm). El ANOVA no detectó diferencias significativas globales,

aunque se observaron disparidades morfológicas y de distribución en el análisis por tamices. La combinación de molinos optimizó la distribución granulométrica,

superando las limitaciones de cada equipo por separado.

Palabras clave.

Distribución granulométrica, Molino de martillo, Molino de bolas, Molienda de maíz, Molienda de soja.

1. Introduction

Particle size reduction is a process implemented in various

industries, which consists of reducing the physical

dimension of solid materials through the application of

mechanical forces. This process is essential in operations

such as mixing, drying, sintering and chemical reactions,

where particle size can influence the speed and uniformity

of the process. Commonly used equipment for size

reduction include ball mills, hammer mills, jaw crushers,

and roller mills. The choice of the right equipment depends

on the properties of the material and the desired particle

size, being a critical factor for the optimization of

industrial processes. [1]

In addition, particle distribution plays a key role, as it

directly affects the quality and properties of the final

product, such as flow, compaction and dissolution. Particle

size analysis is an essential technique for evaluating the

University of Guayaquil / stefanie.bonillab@ug.edu.ec ; https://orcid.org/0000-0002-9391-3698, Guayaquil; Ecuador.

Independent Researcher / fernandoanoblex18@gmail.com; https://orcid.org/0009-0005-1898-8373, Guayaquil; Ecuador.

Independent Researcher / ivanalejo17@gmail.com;https://orcid.org/0009-0008-9193-0524, Guayaquil; Ecuador.

University of Guayaquil / carlos.valdiviezor@ug.edu.ec; https://orcid.org/0000-0002-6550-975, Guayaquil, Ecuador.

particle size distribution in a pulverized material, and the

sieve is one of the most widely used pieces of equipment

for this purpose. Accuracy in particle classification is vital

to ensure product consistency. [2]

Grinding and size reduction not only increase the specific

surface area of materials, but also improve their reactivity

and facilitate downstream processes such as dissolution,

extraction of compounds of interest, and homogenization

into mixtures. In the food industry, for example, proper

control of particle size helps to optimize the texture,

solubility and bioavailability of nutrients, while in the

pharmaceutical industry particle size uniformity is key to

ensuring the dosage and controlled release of active

ingredients. [3]

In the field of construction and mining materials, the

efficiency of comminution equipment, such as ball mills

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 103

and hammer mills, has a direct impact on energy

consumption and operating costs. It is estimated that up to

50% of the total energy used in a mineral processing plant

corresponds to milling operations, which makes equipment

selection and particle size optimization strategic factors for

process sustainability. In addition, excessive particle

reduction can lead to material losses due to fines

formation, affecting the overall efficiency of the

system.[4]

The hammer mill is one of the most widely used equipment

for grain size reduction due to its simplicity of design, low

cost, and high processing capacity. Its operating principle

is based on the repeated impact of rotary hammers on the

particles, which generates rapid fractures and produces

materials with a relatively heterogeneous particle size.

This type of mill is widely used in the food and feed

industry, as it allows grains such as corn, wheat and

soybeans to be processed efficiently, although it has the

disadvantage of generating a higher content of fines and

dust. [5]

On the other hand, the ball mill operates under the principle

of impact and friction, where spheres of steel or other

grinding material rotate inside a cylindrical drum, causing

the gradual reduction of the particle size. Unlike the

hammer mill, this equipment allows for a more controlled

and finer distribution of particles, with less variability in

size. Ball mills are widely used in the mining, ceramics,

and pharmaceutical industries, as well as in the research of

new materials, although they require higher energy

consumption and longer operating times compared to

hammer mills. [6]

Particle size analysis by sieving, laser diffraction or other

modern methods is used as a quality control tool to

establish the size distribution in the processed products.

The sieving technique, although traditional, is still one of

the most widely used due to its low cost, simplicity and

reproducibility compared to more sophisticated methods.

The information obtained from these analyses makes it

possible to establish correlations between the distribution

of particles and the behaviour of the material in subsequent

processes, guaranteeing the uniformity of the final product

and contributing to the optimisation of the production

chain. [7]

The choice between a hammer mill and a ball mill depends

largely on the material to be processed and the desired

properties in the final product. For grains, the hammer mill

is preferred for its speed and efficiency in large volumes,

while the ball mill is more appropriate when fine, uniform

grinding is required. Both pieces of equipment play a

fundamental role in the optimization of industrial

processes, and their comparison from the perspective of

particle size distribution allows us to identify competitive

advantages and areas for improvement in the reduction of

particle size. In this context, it is pertinent to highlight the

importance of milling in massively used grains such as

corn and soybeans, whose processing not only responds to

industrial purposes, but also to the optimization of the

nutritional and functional quality of the derived products.

[8][9]

Despite the widespread use of hammer and ball mills in

different industries, there are still gaps in the comparative

understanding of their efficiency in the size reduction and

in the final particle size distribution of grains such as corn

and soybeans. While both equipment serves similar

functions, differences in their operating principle, energy

consumption, and product uniformity can significantly

influence the quality and utilization of processed grains.

Recent studies have highlighted that grinding parameters,

such as rotation speed, ball loading or screen opening, have

a direct impact on the distribution of particles and the

nutritional quality of the final product. Comparative

research has shown that hammer mills tend to generate

more irregular particles and a higher content of fines, while

ball mills produce more homogeneous distributions,

although with higher energy consumption and operating

time. However, most of these studies have focused on

individual grains or specific experimental conditions, so a

more comprehensive analysis is required that relates both

equipment under controlled and comparable conditions. In

this way, the present research seeks to provide quantitative

and updated evidence that allows to guide technical and

economic decisions in the processing of corn and

soybeans, strengthening the scientific basis for the

selection of the most efficient milling system.[10][11]

In the case of grains such as corn and soybeans, which are

widely used in the food and feed industry, milling plays a

key role in improving their functional and nutritional

properties. In corn, the control of particle size influences

the digestibility of starch and the quality of derived

products such as flour and cereals, while in soybeans it

determines the availability of proteins and lipids, in

addition to facilitating their incorporation into balanced

formulations for animal feed. Studies have shown that the

adequate reduction of the particle size in these grains not

only optimizes the performance of the extraction processes

and digestibility, but also impacts the [12]energy efficiency of

the milling and in the final quality of the product. [13]

Previous studies have specific limitations that require

attention. For example, research showed that, in corn

milling, the specific energy required varies considerably

according to the fraction of the material (grain, stubble,

rope), which suggests that the data cannot be directly

extrapolated to processed commercial grains. Another

study showed that the combination of mills (hammers +

rollers) improves the uniformity of particle distribution,

but it does not directly compare hammer mills vs ball mills

in grains such as corn or soybeans. In the field of ball

milling, a study looked at how the diameter of the medium

affects milling efficiency, but in mineralization, not in

agricultural grains, which leaves a gap in knowledge

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 104

applicable to the food sector. Consequently, there is a lack

of a direct comparison, under controlled conditions of food

grains (corn and soybeans), between hammer mills and

ball mills, which simultaneously quantifies particle size

uniformity, energy consumption and their link with

nutritional or procedural functionality. This gap gives

relevance and urgency to the present research, aimed at

ensuring a solid technical selection of the most suitable

grinding system for its industrial application.[14][15][16]

Recent studies indicate that hammer milling can represent

up to 50% of a power plant's total electricity consumption.

On the other hand, research in biomass shows that the

specific energy required for size reduction can vary

between 35–65 kJ/kg, depending on the type of material

and the grinding conditions. In addition, analyses with

empirical models indicate that the energy required in ball

mills can vary between [17]~3–12 kW·h·t⁻¹ depending on

the hardness and desired product size. Therefore, improper

selection of the type of mill not only affects the quality of

the grind and the particle size uniformity, but can also

considerably increase the [18]Operating costs and the

Energy consumption, impacting the viability and

competitiveness of the industrial process.

Therefore, the objective of this study is to perform a

comparative analysis of the particle size distribution of

ground grains in ball mills and hammer mills, considering

their application in the processing of raw materials such as

corn and soybeans. This analysis seeks to establish

relationships between the type of equipment, the grinding

conditions and the uniformity of the particles obtained, in

order to provide technical criteria that guide the selection

of the size reduction system based on the efficiency and

quality of the final product.

1.1. Grinding

Grinding is a unitary operation which is responsible for

reducing the particle size to achieve a size required for a

specific process, thus increasing the contact surface of the

material for greater efficiency in the industrial process.

This reduction is carried out by dividing or fractionating

the sample by mechanical means until a required size can

be reached.

For Chemical Engineering it is essential to

understand the laws that govern disintegration in

relation to energy consumption (time), the

characteristics of the matter and the type of

machines to be used, this demonstrates the study

based on deductions and empirical observations.

[1]

1.1.1. Types of Grinding

Different types of mills such as ball mill and hammer mill,

have different mechanisms of action and efficiency, ball

mills are efficient for fixed grinding and hammer mill is

more for fragile materials. [19]

1.2. Sieving

The sieving method involves using a series of sieves with

different openings to separate soil particles according to

their size [20]

1.3. Granulometric analysis by sieving

It is the separation in size of a collection of solid particles

according to a particle size scale. This separation is carried

out with sieves placed in series, so that the sifting of the

first sieve is the feed of the second and so on. [21]

Feeding to the sieve (F): It is the total mass that arrives at

the sieve to be separated or classified.

Retained (R): It is the mass that remains on the surface of

the sieve.

Sifting (C): It is the mass that passes through the openings

of the sieve, that is, that passes through its surface.

1.4. Particle Size Distribution

Particle size distribution describes the proportion of

different particle sizes present in a sample. It is essential to

characterize the behavior of materials in various industrial

processes. [22]

1.5. Granulometric curves

Particle size curves are graphical representations that show

the particle size distribution in a sample. These curves are

essential to understand the distribution and predictability

of material behavior. [23]

2. Materials and methods.

Grinding tests were carried out with two types of grains

(corn and soybeans) to evaluate the granulometry that

could be obtained from grinding in the ball and hammer

mills. We worked with a hammer mill model RBN rose,

with 20 hammers with cast bar and 4 shafts, as well as with

a tubular, discontinuous steel ball mill, single-chamber,

with grate discharge. The calibration of the grinding and

sieving equipment was carried out prior to the tests,

verifying that all the components: chamber, grinding

bodies, sieves and mesh were within their dimensional and

mechanical specifications, and ensuring reproducible

conditions between replicates. A set of sieves certified to

standards equivalent to ASTM E11 / ISO 3310 were used

for sieving, and the uniformity of the openings was

checked with mesh calibration methods according to

recommended procedures in the literature to ensure

accuracy and minimize classification errors. The following

are the case studies evaluated:

Case 1.Hammer mill grinding and sieving, fig. 1.

Case 2.Ball mill grinding and sieving, fig. 2.

Case 3.Grinding in a hammer mill, followed by the ball

mill and sieving, fig. 3.

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 105

Once the three cases were carried out, a microscope

(Digital Microscope USB) was used to examine the

geometry obtained in each type of grinding.

The sampling protocol was established following criteria

of representativeness and homogeneity recommended for

granulometry studies in agricultural matrices. For each

milling treatment, samples were collected immediately

after unloading the equipment, employing the manual

quartering method to reduce the volume and ensure that the

fraction analyzed maintained the original batch

distribution. This procedure is widely used in milling

studies due to its effectiveness in minimizing size and

density segregation biases, especially in grains such as

corn and soybeans. Recent research emphasizes that

correct homogenization and reduction of batch size is

essential to ensure the reproducibility of the particle size

distribution, since variations in the sampling protocol can

generate differences of up to 15% in the percentage

retained by sieve in impact or compression grinding

systems. In addition, comparative studies in agricultural

milling recommend using between 200 and 500 g as the

minimum analytical mass to avoid losses of fine fractions

and ensure sufficient representativeness, which was

considered in the present work.[24]

For the representation and analysis of the collected data,

granulometric distribution graphs of particle size

distribution were used. These plots allow the relationship

between particle size and the cumulative percentage of the

sieved material to be visualized, providing a quantitative

and comparative understanding of the particle size

distribution.

Finally, the analysis of variance (ANOVA) of a single

factor was applied using the Analysis Toolpak

complement, which allowed the evaluation of significant

differences between treatments.

Fig. 1. Case 1. Hammer mill grinding and sieving.

Fig. 2. Case 2. Ball mill grinding and sieving

Fig. 3. Case 3. Hammer mill grinding, followed by ball mill and sieving

2.1. Raw material

4540 g of corn and soybeans were used, 2270 g of each

grain. Corn with an average diameter of 0.78 mm and grain

density of 0.75 g/cm3, soybeans with an average diameter

of 0.57 mm and grain density of 0.85 g/cm3.

The moisture of the grain is a determining variable in the

efficiency of the milling, as it modifies its hardness, its

mechanical response and the resulting granulometry. To

maintain experimental stability, the grains were kept in the

same batch and stored in a dry environment at a controlled

temperature of 22–24 °C, conditions that minimize

hygroscopic variation and preserve the physical properties

of the material. This approach coincides with

recommendations from the literature, which highlight that

the simultaneous control of temperature and environmental

conditions avoids fluctuations in the internal humidity of

the grain and, therefore, in its behavior during the

comminution process. This ensures that the observed

differences in grain size mainly reflect the performance of

the grinding equipment.[25]

2.2. Ball Mill

A ball mill was used, with a ball load configured as

indicated in table 1:

Table 1. Configuration of grinding bodies.

Grinding

bodies

Average

diameter

(cm)

Total Weight

(g)

% Weight

Small

2.46

5447

18.43

Medium

2.97

9286

31.43

Large

3.89

14815

50.14

Small

2.46

5447

18.43

Source: Bonilla, et al, 2024

The total weight of grinding bodies was 29548 g.

The average weight and equivalent diameter of grinding

bodies were determined with the following equations:

 





[1]

Average weight= 126.27 g

  



















[2]

Equivalent diameter=. 3.12 cm

2.2.1. Parameter Calculations for Ball Mill

The ball mill was fed with 2270 g of each grain (corn and

soybean) and the working parameters were determined by

the following design equations:

Degree of filling (f):

 





 [3]

Weight of Grinding Body Load (Q):

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 106

 









 [4]

Where:

D: Inner diameter of the ball mill, m.

Li: length of the mill, m.

Yq: Equivalent weight of grinding bodies, t/m3

Critical mill speed (nc):

 







[5]

Where:

D: Inner diameter of the ball mill, m

   [6]

Where:

K: Percentage of Critical Speed (75%)

NC: Critical Speed, RPM

Power of the Ball Mill Motor (N):

   [7]

Where:

c: power consumption factor, dimensionless

D: inner diameter of the mill, m

Q: Loading weight of ball mill, t

N: Mill Operation Speed, RPM

Specific energy consumption (CEE):

 





[8]

Where:

N: mill power, kW

P: Ball mill production, t

2.3. Hammer Mill

The grains were placed in the feed mouth, passed through

the hammers in a period of 4 minutes and received at the

unloading, for weighing and sieving.

2.4. Sieving

The sieves were placed in column in an ascending manner

according to the sieve number, which means that the sieve

with the highest number will receive the finest material.

The column of sieves was placed in the vibrating machine

for one minute and then each sieve was weighed and the

weight of the retained sieve was collected.

The study process flowchart is shown below in Figure 4.

Fig.4. General flow diagram of the process.

2.5. Sieves

The column of sieves used in this study is shown in table

2 with their respective characteristics as follows:

Table 2. Sieve Classification

Mesh Number

Mesh Opening (mm)

Sieve Weight

(g)

4,00

387

3,35

383

2,36

372

1,70

358

1,18

309

1,00

303

0,85

294

0,60

289

0,30

255

0,212

250

Base

270

Source: Bonilla, et al, 2024

3. Analysis and Interpretation of Results.

3.1. Grinding Operating Conditions

The following table 3 presents the working conditions in

cases 2 and 3 with both corn and soybean grains.

Table 3. Operating conditions in milling

Parameter

Case 2

Case 3

Grains

Corn

Soybeans

Corn

Soybeans

Temperature

(C)

f (%)

14,13

14,8

14,1

16,51

Q (t)

0,0280

0,0293

0,0279

0,0327

nc (rpm)

68,61

n (rpm) @ 75%

N (Hp)

0,194

0,203

0,194

0,203

CEE (kW h / t)

80,61

83,17

80,17

92,70

Source: Bonilla, et al, 2024

Table 3 indicates the conditions used in the ball mill in

cases 2 and 3, for both grains worked at room temperature.

The degree of filling (f) of case 2 with corn presents

14.13% and soybeans 14.8% respectively. The denser the

grain, the higher the degree of filling. In case 3 with corn,

Weighing the grains

Grinding

Sifting

Heavy

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 107

it presents 14.1% and soybeans 16.51%, these results

depend on factors such as the volume of loading of

grinding bodies, volume of raw material and volume of the

mill. Therefore, corn flour, being denser, occupies less

volume in the mill in contrast to soybean meal that

occupies a greater volume in it, for this reason a greater

load of grinding bodies was required.

It was necessary to determine the critical speed (nc) at

which the mill operates in order not to exceed it since it

will cause the centrifugal force to equal the force of gravity

and the grinding bodies would not descend to the grinding.

The specific energy consumption (CEE) calculated in case

2 with corn had a consumption of 80.61 kW hr/t for each

ton processed, in contrast soybeans increased to 83.7 kW

hr/t consumed for each ton processed, this is due to the fact

that they worked with a higher load of grinding bodies for

this grain so the mill had to consume more energy than

with corn. finally, case 3, with corn, presented a

consumption of 80.17 kW hr/t for each ton produced and

soybeans consumed 92.70 kW hr/t for each ton produced,

the difference in specific energy consumption (CEE)

between the grains in case 3 is due to the density of the

soybean grain. In other words, the higher the degree of

filling, the higher the specific energy consumption. [26]

3.2. Mill yields

Table 4 compares the percentage of yield of the mills in the

three case studies with corn and soybeans.

Case 1. Hammer mill.

The hammer mill presented a higher yield when processing

corn grain compared to soybeans, with a difference of

3.5%. This variation is mainly attributed to the difference

in densities of the two

grains. Corn, being less dense, makes it easier to grind

compared to soybeans.

Case 2. Ball mill.

In the ball mill, the highest yield was obtained when

grinding both grains compared to the three cases analyzed,

due to the ability of the ball mill to process almost all the

grain fed with losses

attributable to incrustations in the shielding and the mill

cover.

Case 3. Hammer mill + ball mill.

The combination of the mills presents a difference between

corn and soybeans of 5.47%, in turn it presents a higher

percentage of loss compared to cases 1 and 2, this is

because the grain goes through two milling processes.

The hammer mill had a superior yield with corn grain

compared to soybeans, due to their differences in densities.

The ball mill offers a similar performance in both cases,

because it does not present a major loss at the time of

processing.

Table 4. Mill yields.

Source: Bonilla, et al, 2024

3.3. Comparative particle size analysis

3.3.1. Case 1. Hammer Mill with Corn

Figure 5, showing the distribution of particles, which

covers a range from 2.33 to 3.7 millimeters, indicates a

lack of uniformity in the reduction of size. While the

median diameter (D50) is approximately 2.8-2.9

millimeters.

Fig.5. Particle size curve Case 1 (maize)

The data in Table 5 indicate a heterogeneous distribution

with a massive concentration in the middle range. Sieve

number 12 (2.03 mm/2030 microns) retains 38.94% of the

total material, representing the maximum point of

distribution. This indicates that the milling process

predominantly generates medium-sized particles.

Although the coarse fraction is adequately minimized

(only 0.47% above 3.68 mm), the low proportion of fines

(10.71% under 0.45 mm) suggests inefficiencies in the

fracture mechanism, possibly related to rotor speed,

residence time, or grain moisture.

Table 5. Experimental data % retained (maize)

Mesh

Number

Average

particle size

(mm)

Average

particle size

(microns)

% Retained

4,00

4000

0,06

3,35

3680

0,41

2,36

2860

11,87

1,70

2030

38,94

1,18

1440

15,77

1,00

1090

7,04

0,85

930

5,18

0,60

730

8,32

Yield (%)

Case #1

Case #2

Raw

material

Hammer

Mill

Ball Mill

Hammer Mill +

Ball Mill

Corn

77,05

96,34

74,89

Soy

73,57

96,17

69,47

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 108

0,30

450

8,50

0,212

260

2,21

Source: Bonilla, et al, 2024

3.3.2. Case 1. Hammer mill (soybean)

Figure 6 shows the distribution of particles in a wide range

of sizes from approximately 2.3 mm to 3.7 mm, which

indicates a heterogeneous grinding with simultaneous

presence of fine and coarse particles. The D50 (medium

size) is located around 2.9 mm, the point where 50% of the

material is thinner and 50% thicker, a value that represents

the characteristic size of the final product. The gentle slope

of the accumulation curve suggests an extended

distribution with significant dispersion in particle sizes. It

is observed that approximately 30-35% of the material has

sizes greater than 3.1 mm (coarse fraction), while only

about 15-20% is below 2.5 mm (fine fraction), evidencing

an imbalance towards larger particles. The accumulation

of retained material indicates that there is a proportion of

particles in the medium range (2.7-3.1 mm), representing

between 40-50% of the total. This distribution suggests

that the grinding process generates an excess of

intermediate-sized particles, possibly due to operating

parameters such as inadequate rotor speed, insufficient

residence time, hammer wear, or excessive sieve opening.

Fig.6. Particle size curve Case 1 (soybean)

The data in Table 6 indicate that the new sieves have a

higher percentage of retention for particles in the range of

3680 to 2860 microns, which is associated with the

presence of larger particles of soybeans. presence of larger

particles of soybeans. In addition, a decrease in the

retention percentage is observed for intermediate and small

sizes. This trend indicates that the

Table 6. Experimental data % retained (soy)

Mesh

Number

Average

particle size

(mm)

Average

particle size

(microns)

% Retained

4,00

4000

2,66

3,35

3680

40,49

2,36

2860

37,97

1,70

2030

5,87

1,18

1440

3,99

1,00

1090

1,96

0,85

930

1,33

0,60

730

2,24

0,30

450

2,80

0,212

260

0,63

Source: Bonilla, et al, 2024

3.3.3. Case 2. Ball Mill (Corn)

The particle size distribution obtained from the corn mill

in the ball mill indicates a wide dispersion of particle sizes.

The results in Figure 7 show that the D10 is about 0.48

mm, which means that only 10% of the material is smaller

than this value, while the D50 is 1.38 mm, indicating that

half of the material is below this size. The 1.96 mm D90

reveals that 90% of the material is smaller than this value.

These indicators allow us to infer that the milling produces

particles in a considerable range, mostly concentrated

between 0.5 mm and 2 mm.

The cumulative distribution shows that the ground material

has a significant proportion of coarse particles (>2 mm),

approximately 8% of the total, which indicates that the ball

mill does not achieve a completely fine grind for all the

processed material. The dispersion of particle size is

relatively high, as confirmed by the uniformity index

(D90/D10 ≈ 4.08), indicating that there is a considerable

mixture of fine and coarse particles. This characteristic is

common in grinding done in ball mills.

Fig.7. Cumulative particle size curve Case 2 (maize)

3.3.4. Case 2. Ball mill (soybean)

The particle size distribution of the ground soybeans in

Figure 8 shows that the material is mostly coarse: the

representative parameters are D10 ≈ 1.59 mm, D50 ≈ 3.53

mm and D90 ≈ 3.94 mm. This indicates that 10% of the

particles are smaller than 1.59 mm, the median is 3.53 mm

(half of the material is finer than this value) and that 90%

is finer than 3.94 mm. These values place most of the mass

in the range ~1.6–4.0 mm, with the median close to 3.5

mm. The pitch curve (the through-% in the meshes) has a

steep slope in the span between approximately 2.86 mm

(18.9% through) and 4.00 mm (97.34% through-screen),

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 109

which means that a significant fraction of the material

passes through large meshes and that the "cut-off zone" of

the distribution is centered around 3–4 mm. The fraction

of fines (<1 mm) is very small (through-% values at 0.45

mm and 0.26 mm are 0.70% and 0.07% respectively),

therefore, the fines are practically insignificant in the final

product. The dispersion of the distribution can be

quantified with the uniformity index D90/D10 ≈ 3.94 / 1.59

≈ 2.48, indicating a relatively narrow and fairly uniform

distribution around coarse sizes (less dispersion than a very

wide distribution). In other words, most of the material is

clustered in a fairly compact range (mostly between ~1.6

and ~4 mm), without large tails of very fine particles or a

very heterogeneous mix of sizes. From an operational and

product quality point of view, this granulometry suggests

that the milling process is producing a suitable product

when looking for coarse or intermediate meal/particles

(e.g. for certain industrial uses or animal feed). If the goal

was to obtain finer fractions or increase the proportion of

particles <1 mm, it would be necessary to intervene in the

process (more grinding time, higher impact energy, adjust

load and size of media, or use sorting and recirculation).

Fig.8. Cumulative particle size curve Case 2 (soybeans)

3.3.5. Case 3. Grinding Hammer Mill + Ball Mill

(Corn)

Combined grinding (hammer + balls) generates an

intermediate-sized product, with a D50 ≈ 1.08 mm, which

means that half of the material is in the range of particles

close to 1 mm. The D10 ≈ 0.60 mm indicates the finest

fraction of the material, while the D90 ≈ 2.60 mm shows

that 90% of the material is below this size, with a wide

range of distribution.

The curve accumulated in Figure 9 shows that most of the

material is concentrated between 0.6 and 2.6 mm, with a

very low fraction of fines (<0.5 mm) (<1%). This indicates

that the milling is efficient to produce medium

granulometry, without excess powders, a favorable

characteristic for uses in animal feed and processes that

require an adequate flow without caking.

Fig.9. Cumulative particle size curve Case 3 (maize)

The uniformity index, calculated as D90/D10 ≈ 4.3,

indicates a wide distribution, with coexistence of fine and

coarse particles. This amplitude may be associated with the

combination of grinding technologies: the hammer mill

breaks more irregularly, and the ball mill refines, but

maintains some dispersion.

3.3.6. Case 3. Grinding Hammer Mill + Ball Mill

(soybean)

The resulting particle size distribution shows in Figure 10,

a clear concentration in the range ≈0.9–1.5 mm, with D50

≈ 1.35 mm, which means that half of the mass is below that

size. The sieving data indicate that the largest quantities

were retained in 1.25 mm and 1.00 mm meshes (498 g and

588 g respectively), underlining that the "critical mass" of

the material is around these sizes and that the process

produced a product of intermediate particle size. The

fraction of fines is very small: only 0.94% passes the 0.315

mm mesh and the accumulated through-that mesh is

0.94%, so particles smaller than ≈0.5 mm are practically

insignificant. This means low dust generation and greater

handling and transport facilities (less aerial dispersion and

fewer problems of caking by fines), a positive aspect for

logistics and for subsequent processes that benefit from

less dusty material. The dispersion of the distribution can

be quantified with the uniformity index D90/D10 ≈ 2.05 /

0.77 ≈ 2.66, a value that indicates a relatively narrow and

homogeneous distribution around the median size. In

practice this means that most of the material is grouped in

a limited range (low fine tail and moderate coarse tail),

which is desirable when looking for reproducibility in

rheological, particle size and process properties (mixing,

extrusion, pelletizing).

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 110

Fig.10. Cumulative particle size curve Case 3 (soybean)

From an operational point of view, these results suggest

that the hammer + ball mill combination produced a

sufficient reduction for applications requiring intermediate

particles.

3.3.7. Comparison of the percentage of retention in

each case in the sieves with the smallest and

largest aperture.

Table 7 shows a comparison between the three case studies

with respect to the percentage of retained obtained in the

meshes with the largest aperture (4mm) and smallest

aperture (0.121 mm), as well as the retention in the base

(<0.212 mm).

Table 7. Comparison of the percentage of detainees.

No.

Mesh

Openness

(mm)

Case 1

(%)

Case 2

(%)

Case 3

(%)

corn

Soy

corn

Soy

corn

Soy

0,06

2,66

2,47

35,01

0,23

0,77

0,212

2,21

0,63

1,12

1,28

1,93

0,28

Base

< 0.212

1,69

0,07

0,04

0,09

0,12

0,14

Source: Bonilla, et al, 2024

The highest percentage of retention in soybean milling was

obtained in case 2 with 35.01%, due to the density of the

grain compared to corn. In case 3, a decrease in retention

is observed since, having a previous grinding in the

hammer mill, smaller particles could be obtained. In the

three case studies, it was possible to obtain smaller

particles, however, the retention percentages do not exceed

3%. The highest accumulation of retained was obtained in

corn milling in all three cases, due to the lower density of

the grain. The soybean milling in the three case studies did

not reach 1% in those retained in the particles less than

0.212 mm. However, in the maize milling, only in case 1

could 1.69% of retained with particle size less than

0.212mm be obtained.

3.3.8. Microscopic analysis of corn and soybean meal

The microscopic analysis of corn flour shows in table 8,

differences in the shape of the particles produced by the

different case studies.

Case 1 (Hammer Mill): The particles have a rectangular

shape, this suggests that the hammer mill's impact

mechanism tends to fragment the particles into angular

shapes. Case 2 (Ball Mill). The particles have an oval

shape, indicating that the abrasion process in the ball mill

tends to round the particles. Case 3 (Hammer Mill + Ball

Mill): An intermediate shape between rectangular and oval

is observed. This may be due to the combination of both

grinding processes, which produces a mixture of particle

forms. [27]

Table 8. Microscopic analysis of corn flour

Case 1

Case 2

Case 3

Particle size: <

0.212 mm Particle

shape: Rectangular.

Particle size: <

0.212 mm Particle

shape: oval.

Particle size: <

0.212 mm Particle

shape: Between

rectangular and

oval.

Source: Bonilla, et al, 2024

3.3.9. Microscopic analysis of soybean meal

Microscopic analysis of soybean meal shows the following

observations in Table 9.

Case 1 (Hammer Mill): The particles have a rectangular

shape, the direct impact of the hammer mill produces

angular particles similar to those observed in corn. Case 2

(Ball Mill). Particles have an ellipsoid shape. The abrasion

process of the ball mill rounds the particles, generating this

shape by the mill's impact method and the characteristics

of the grain. Case 3 (Hammer Mill + Ball Mill): A mixture

of rectangular and ellipsoidal shapes is observed. This can

be a result of process combinations, where pre-grinding in

the hammer mill followed by abrasion in the ball mill

produces a range of various shapes.

Table 9. Microscopic analysis of soybean meal

Case 1

Case 2

Case 3

Particle size: <

0.212 mm Particle

shape: Rectangular.

Particle size: <

0.212 mm Particle

shape:

Ellipsoid

Particle size: <

0.212 mm Particle

shape: Between

rectangular and

ellipsoid.

Source: Bonilla, et al, 2024

3.3.10. Statistical analysis

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 111

An analysis of variance was performed for a factor that

allowed in each case to determine if there are significant

differences between the results.

3.3.10.1. Case 1. Hammer Mill

Table 10 shows the analyzed data on the weight of the

retained in the sieves of corn and soybean grains.

Null hypothesis: The granulometric analysis does not show

significant differences between the sieves used. (p>0.05)

Alternative hypothesis: The granulometric analysis shows

significant differences between the sieves used. (p<0.05)

Table 10. Weights of Retained in Sieves Case 1

Mesh (mm)

2,36

1,70

1,18

1,00

0,85

0,60

0,30

0,212

Corn (g)

472

701

296

2.21

Soybeans

(g)

543

2.93

Critical Value F

4,39

Probability

0,33

Source: Bonilla, et al, 2024

The one-factor analysis of variance (ANOVA) applied to

the particle size distributions of corn and soybeans shows

that there is no statistically significant difference between

both milled products, evidenced by a p-value of 0.33, much

higher than the usual significance level of 0.05, which

indicates that the variations observed in the retained

weights in each mesh can be attributed to the randomness

of the process rather than to an actual effect of the type of

material. However, it may mask relevant practical

differences in milling patterns, such as the marked

concentration of soybeans in the 2.36 mm (543 g) mesh

versus a more uniform distribution of corn, suggesting that,

although globally similar, the fracture mechanisms and

breakdown characteristics of each material could be

influenced by factors not captured by this univariate

analysis.

3.3.10.2. Case 2. Ball Mill

Table 11 shows the analyzed data on the weight of the

retained in the sieves of corn and soybeans.

Null hypothesis: The granulometric analysis does not show

significant differences between the sieves used. (p>0.05)

Alternative hypothesis: The granulometric analysis shows

significant differences between the sieves used. (p<0.05)

Table 11. Weights of Retained in Sieves Case 2

Mesh (mm)

2,36

1,70

1,18

1,00

0,85

0,60

0,30

0,212

Corn (g)

482

511

447

202

153

217

1.63

Soybeans

(g)

762

338

192

127

1.28

Critical Value F

4,49

Probability

0,61

Source: Bonilla, et al, 2024

The analysis of variance (ANOVA) of a factor applied to

the particle size distribution data of corn and soybeans

ground in a ball mill indicates that there is no statistically

significant difference between the distributions of both

grains, evidenced by a p-value of 0.612, much higher than

the significance level of 0.05, and an F-value (0.267) that

does not exceed the critical value (4,494). Although the

means of the retained weights differ numerically (maize:

236.74 g, soybean: 183.25 g), the high variance within

each group (38,378.68 for maize and 58,008.20 for

soybean) suggests that the dispersion of the data in each

mesh is considerable, masking possible specific

differences by particle size. This could be due to the

heterogeneous nature of grinding in ball mills, where

factors such as hardness, humidity or residence time

generate variability that the global ANOVA fails to detect,

recommending an analysis by specific particle size

fractions to identify practical differences in the process.

3.3.10.3. Case 3. Hammer Mill + Ball Mill

Table 12 shows the analyzed data on the weight of the

retained in the sieves of corn and soybeans.

Null hypothesis: The granulometric analysis does not show

significant differences between the sieves used. (p>0.05)

Alternative hypothesis: The granulometric analysis shows

significant differences between the sieves used. (p<0.05)

Table 12. Weights of Retained in Sieves Case 3.

Mesh (mm)

2,36

1,70

1,18

1,00

0,85

0,60

0,30

0,212

Corn (g)

1.9

Soybean

s (g)

0.2

Critical Value F

4,49

Probability

0,63

Source: Bonilla, et al, 2024

The analysis of variance (ANOVA) of one factor applied

to the particle size data of case 3, where corn and soybeans

were processed by a hammer mill followed by a ball mill,

indicates that there is no statistically significant difference

between the particle size distributions of both grains. This

is supported by a p-value of 0.637, much higher than the

significance level of 0.05, and an F-value of 0.231 that

does not exceed the critical value of 4.494. Although the

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 112

mean retained weight differ (maize: 185.77 g, soybean:

155.14 g), the high variability within each group (variances

of 24,235.87 for maize and 12,286.76 for soybeans)

suggests that the observed differences may be due to the

natural dispersion of the milling process and not to the type

of grain. This result reflects that the grinding sequence

(hammer + balls) homogenizes the distributions to the

point of eliminating significant differences, possibly due to

the combination of fracture mechanisms (impact and

abrasion) that compensate for the individual properties of

each material. However, an analysis by specific fractions

could reveal differential behaviors in particular size

ranges, not captured by the global ANOVA.

3.3.10.4. Diameters Comparisons in Corn

Table 13 shows the data analyzed for diameters D10, 50

and 90 for maize in its three case studies.

Null hypothesis: There are no significant differences

between the values of D10, D50 and D90 in the particle

size distribution of maize. (p>0.05)

Alternative hypothesis: At least one of the percentiles

(D10, D50 or D90) differs significantly from the rest in the

particle size distribution of maize. (p<0.05)

Table 13. Through-the-crop accumulation at D10, 50 and 90 for corn

Parameter

Case 1

Case 2

Case 3

D10 (mm)

3,39

0,48

0,60

D50 (mm)

2,80

1,38

1,08

D90 (mm)

0,30

1,96

2,60

Critical Value F

0,03

Probability

0,97

Source: Bonilla, et al, 2024

The analysis of variance shows that there are no significant

differences between the values of D10, D50 and D90 for

corn grain (p > 0.05). This indicates that, considering the

three milling treatments, the variation between the

characteristic particle size distributions (size percentiles) is

statistically similar. In other words, milling generates

particle sizes that, although different in numerical value,

do not present sufficient variability for the differences

between percentiles to be statistically detectable in corn.

3.3.10.5. Soybean Diameter Comparisons

Table 14 shows the data analyzed for diameters D10, 50

and 90 for maize in its three case studies.

Null hypothesis: There are no significant differences

between the values of D10, D50 and D90 in the particle

size distribution of soybeans. (p>0.05)

Alternate hypothesis: At least one of the percentiles (D10,

D50 or D90) differs significantly from the rest in the

particle size distribution of soybeans. (p<0.05)

Table 14 Accumulation of the through-a-basket in D10, 50 and 90 for

soybeans

Parameter

Case 1

Case 2

Case 3

D10 (mm)

1.,8

1.,9

0,70

D50 (mm)

2,90

3,53

1,35

D90 (mm)

3,98

3,94

2,05

Critical Value F

4,04

Probability

0,07

Source: Bonilla, et al, 2024

In the case of soybeans, there is a trend towards differences

between percentiles (p = 0.773), but these do not reach

statistical significance at the conventional level of α = 0.05.

This suggests that soybeans have greater particle size

variation than maize between D10, D50 and D90, probably

due to their more fragile internal structure and greater

heterogeneity in the fracture. However, statistical evidence

is not sufficient to affirm significant differences, although

it is possible that with a larger sample the results may be

conclusive.

4.- Discussion

4.1.- Case 1. Hammer Mill

The particle size data reveal significant differences in the

milling behavior between soybeans and corn using new

sieves. Soybeans have a Extremely coarse distribution,

with 81.12% of the material retained in the first three

meshes (≥2.36 mm) and only 18.88% as through material

in mesh 8. In contrast, corn shows a More balanced

distribution, with 87.66% of through-material in mesh 8

and a progressive accumulation in intermediate meshes,

reaching its inflection point (D50) around mesh 12 (1.7

mm). This difference is evident in the values of Smaller

percentage accumulation size, where maize maintains

significantly higher percentages across all meshes,

indicating more efficient and uniform grinding. These

findings are consistent with the grinding theory that states

that the Physical properties of the material determine their

response to the fracture. Soybeans, with a higher oil

content and more flexible cell structure, have greater

resistance to impact fracture, resulting in coarser particles.

Maize, with vitreous endosperm and higher starch content,

fractures more easily, generating a finer and more uniform

distribution. The literature reports that materials with

hardness greater than 45 kg/cm² (such as soybeans) require

more grinding energy and produce coarser distributions,

while cereals such as corn (hardness 25-35 kg/cm²)

respond better to impact milling. The behavior observed in

soybeans, with 81.12% of retained accumulated in mesh 8,

It coincides with previous studies that report low milling

efficiency in oilseed legumes due to their ability to absorb

energy without fracturing completely. On the other hand,

the distribution of maize conforms to the Gates-Gaudin-

Schuhman model typical of brittle materials. In other

studies, it reports 72-88% of material under 2 mm for corn,

coinciding with our 87.66% through-mesh. [28] glassy

nature of the endosperm. Just as it confirms that the

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 113

[29]Presence of oil in soybeans (18-22%) it acts as a shock

absorber, reducing the generation of fines by 30-40%

compared to dry materials. In contrast, in other research he

obtains finer distributions (D50 ≈ 1.2 mm) using liquid

nitrogen, suggesting that our conventional conditions limit

efficiency. This shows that real-time adaptive adjustments

can improve soybean uniformity by up to 60%, indicating

optimization potential not explored in our study. Some

studies claim that state-of-the-art mills with variable speed

control can achieve narrower distributions than

conventional equipment, in addition to the incorporation of

current protocols recommends the humidity control

implemented in the studios. These findings coincide with

recent work on feed and PSD behavior in hammer milling,

which shows how soybeans and corn respond differently

to the same screening/screen and how moisture and

composition affect fines generation and nutrient

distribution by fraction. In particular, studies observed

that, after hammer milling, soybeans tend to retain

relatively coarse fractions and that the addition of moisture

significantly modifies both the PSD and energy

consumption, which supports the interpretation that grain

properties (oil, structure) condition the hammer efficiency

and the orientation of the particle size curve.[30][31] [32]

[33] [34]

4.2.- Case 2. Ball Mill

Corn has a Significantly finer distribution than soybeans,

with 91.6% of through-material in mesh 8 (2.36 mm)

compared to only 26.32% in soybeans. The inflection point

(D50) is located approximately at 1.3-1.4 mm for corn

versus 3.0-3.2mm for soybeans, evidencing a marked

difference in grinding efficiency. Soybeans show Fracture

resistance, with 73.68% of retained accumulated in mesh

8, consistent with studies of Chen et al. (2021) on the

ability of oilseed materials to absorb impact energy. Maize,

with its vitreous endosperm, responds better to impact

milling, generating a higher proportion of medium and fine

particles. The results coincide in the higher milling

efficiency of cereals versus oilseeds under similar

conditions, but highlight the need to optimize specific

operating parameters for each material. The interpretation

of the results is based on recent scientific references, such

as the study that explains the fracture resistance of

soybeans due to their high oil content and flexible cell

structure, which coincides with our findings of coarser

particle size distributions compared to corn. Likewise

[28] [35], Other studies they highlight the influence of the

elastic modulus on grinding, supporting the marked

difference in D50 between the two materials. When

comparing our results with the literature, there are

coincidences, who report finer distributions for corn under

similar milling conditions, while discrepancies arise when

contrasted with those who used cryogenic milling in

soybeans and obtained significantly finer distributions,

suggesting that our conventional operating conditions limit

the efficiency of the process. Reviews and experimental

work confirm that the ball mill is very effective in

fractionating and reducing particles when the material is

essentially friable or rich in polysaccharides, but its

performance in dense oilseeds can be limited without

parameter adjustments (time, media, atmosphere). This is

consistent with our results: corn (endosperm) is effectively

reduced, soybeans are not, except by operative changes or

combined treatments.[36][28] [30] [37]

4.3.- Case 3. Hammer Mill + Ball Mill

The granulometric analysis of the Case 3 shows a

noticeably finer and more uniform distribution compared

to previous cases, where the corn presents a D50 ≈ 1.05

mm (near mesh 18) and the soybeans a D50 ≈ 1.25 mm

(between 16-18 meshes), evidencing a more efficient

grinding. This improvement is attributed to the possible

use of hammer mill followed by ball mill, a combination

that according to Some studies in this case It optimizes the

fracture of heterogeneous materials by integrating impact

and abrasion mechanisms. The results coincide with those

who report distributions with D50 between 0.9-1.2 mm for

corn under sequential milling, while discrepancies persist

with respect to soybeans, whose studies indicate D50 >1.5

mm in conventional milling, suggesting that our sequential

process partially mitigates its fracture resistance. Among

the limitations, the absence of humidity control and

unmonitored sieve wear may have affected reproducibility,

while equipment bias underestimates the potential for

cryogenic grinding. These factors highlight the need to

incorporate energy-specific metrics (in future research to

validate the efficiency of the sequential process. So it is

evident that fragmentation methods will impact directly on

the particle size as well as the shape obtained in your flour.

. In recent literature, it has been shown that the use of

sequential stages or multi-stage milling can decrease

energy consumption per unit of reduction (depending on

humidity and target sizes) and stabilize the production of

intermediate/fine particles; In addition, the pre-

fragmentation stage facilitates the action of subsequent

fine grinding and less contact time needed to reach the

target size. [30] [28] [35] [33] [38][39] [40]

Although the results obtained show clear differences

between grains and treatments, there are inherent

limitations to the experimental design and biophysical

factors of the grain that could modulate the milling

efficiency. In the Internal composition (lipid content,

proteins, cell structure) and the grain moisture influence

their susceptibility to fracture. For example, a recent study

shows that when humidity increases, grains (corn, rice and

soybeans) modify their mechanical behavior: they go from

brittle to viscoelastic, changing the fracture force and the

energy required. These variations can alter the

fragmentation under milling, suggesting that intrinsic

factors of the grain beyond the type of mill may condition

the final particle size distribution.[41]

Another hypothesis to explain the lower milling efficiency

in soybeans, compared to corn, is the buffering effect of oil

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 114

content. In soybean milling, it has been reported that higher

lipid contents limit size reduction: the oil reduces

brittleness and favors agglomerates or coarse particles

after impact grinding. Additionally, friction grinding or

fine grinding studies show that mechanical methods affect

grains with a starch-rich matrix differently versus those

with high lipid content, modifying the efficiency, particle

shape and dispersibility. Therefore, the current results

could reflect an interaction between the milling technique

and the biochemical properties of the bean; This alternative

hypothesis deserves to be explored in future work through

characterization of oil, protein and cell structure, and

rigorous humidity control.[42][28]

5.- Conclusions

The hammer mill (Case 1) showed a heterogeneous

distribution, with a D50 ≈ 2.8-2.9 mm for corn and ≈2.9

mm for soybeans, evidencing an inefficient milling in

soybeans, where 81.12% of the material was retained in

≥2.36 mm meshes. In contrast, the ball mill (Case 2)

generated finer distributions for corn (D50 ≈ 1.38 mm), but

was ineffective for soybeans (D50 ≈ 3.53 mm), confirming

that soybeans, due to their higher oil content and flexible

structure, resist fracture by impact and conventional

abrasion.

The combination of mills (Case 3) achieved the most

balanced distribution, with D50 ≈ 1.05 mm for corn and

≈1.25 mm for soybeans, reducing the heterogeneity

observed in the individual cases. This suggests that the

synergy between impact (hammer) and abrasion (balls)

mechanisms mitigates the limitations of each method

separately, although greater dispersion persists in corn

(uniformity index D90/D10 ≈ 4.3) versus soybeans (≈2.66)

The ball mill (Case 2) had the highest specific energy

consumption (CEE): 92.70 kW·h/t for soybeans versus

80.17 kW·h/t for corn in Case 3, associated with the higher

density and resistance of soybeans. However, this high

consumption did not translate into a fine grind for

soybeans, indicating energy inefficiency in oilseed

processing under standard conditions.

The ANOVA applied showed p > 0.05 in all three cases,

indicating "no significant differences" between corn and

soybeans. However, the particle size data reveal critical

operational disparities, such as the concentration of

40.49% of soybeans in the 3.35 mm mesh (Case 1) versus

a more uniform distribution in corn. This exposes the

insensitivity of univariate ANOVA to capture differences

in complex distributions, underscoring the need for

fraction-specific analysis.

The results of the ANOVA for maize (F = 0.0314; p =

0.9692) show a total absence of significant differences

between the particle size parameters D10, D50 and D90

according to the milling method, evidencing a uniform

behavior. In soybeans, although the ANOVA shows F =

4.0430 and p = 0.0773, the difference does not reach

statistical significance, but suggests a tendency to

variation, possibly influenced by its higher oil content and

different mechanical response to fracture.

Microscopic analysis revealed that the hammer mill

generates angular (rectangular) particles, while the ball

mill produces rounded shapes (oval or ellipsoidal). In Case

3, a combination of morphologies was observed, which

affects functional properties such as flow, compaction and

reactivity. This highlights the importance of selecting the

grinding technology according to the desired

characteristics in the final product.

The findings of this study can be applied in the

optimization of milling processes in the feed, flour and

vegetable oil industry, where particle size control directly

influences digestibility, mixture homogeneity and

extrusion efficiency. The sequential grinding configuration

(hammer + balls) is emerging as a viable alternative to

reduce energy consumption and improve particle size

consistency in continuous production lines. It is

recommended to deepen the analysis of the specific milling

energy and its relationship with the moisture and

composition of the grain, as well as to evaluate the effect

of particle size on the nutritional and functional quality of

the final product. In addition, it would be valuable to

explore the implementation of hybrid technologies (such

as cryogenic or ultrasound-assisted milling) to increase the

efficiency and reproducibility of the process, especially in

hard-to-fracture oilseeds.

6.- Author Contributions (Contributor Roles

Taxonomy (CRediT))

1. Conceptualization: Iván Torres

2. Data curation: Alejandro Noblecilla

3. Formal analysis: Alejandro Noblecilla

4. Acquisition of funds: Not applicable

5. Research: Iván Torres, Alejandro Noblecilla

6. Methodology: Iván Torres, Alejandro Noblecilla,

Stefanie Bonilla, Carlos Valdiviezo.

7. Project management: Stefanie Bonilla

8. Resources: Iván Torres, Alejandro Noblecilla

9. Software: Not applicable.

10. Supervision: Stefanie Bonilla

11. Validation: Stefanie Bonilla, Carlos Valdiviezo

12. Visualization: Alejandro Noblecilla

13. Writing - original draft: Iván Torres, Alejandro

Noblecilla

14. Writing - proofreading and editing: Stefanie Bonilla,

Carlos Valdiviezo

7.- References.

[1] L. McCabe and J. Smith, Unit Operations in Chemical

Engineering, Seventh, 2007.

https://iestpcabana.edu.pe/wp-

INQUIDE 
Chemical Engineering & Development 
Journal of Science and Engineering 
Vol.  08 / Nº 01 
e – ISSN: 3028-8533 
ISSN – L: 3028-8533 
 
 
Chemical Engineering & Development 
University of Guayaquil |  Faculty of Chemical Engineering 
Guayaquil – Ecuador 
https://revistas.ug.edu.ec/index.php/iqd 
Email:  inquide@ug.edu.ec  
francisco.duquea@ug.edu.ec 
Pag. 115 
 
content/uploads/2021/11/OPERACIONES-
UNITARIAS-EN-LA-INGENIERIA-QUIMICA.pdf 
[2]  M.  Roostaei,  M.  Soroush,  V.  Arian  and  A.  Ghalambor, 
"Comparison  of  Various  Particle-Size  Distribution-
Measurement  Methods,"  SPE  Reservoir  evaluation  & 
engineering,  vol.  23,  no.  04,  2020. 
https://doi.org/10.2118/199335-PA 
[3]  G. Barbosa-Cánovas, E. Ortega-Rivas and P. Juliano, Food 
Powders:  Physical  Properties,  Processing,  and 
Functionality,  Boston,  MA:  Springer,  2005. 
https://doi.org/10.1007/0-387-27613-0_12  
[4]  A.  Gupta  and  D.  Yan,  Mineral  Processing  Design  and 
Operation,  Western  Australia:  Elsevier.,  2016. 
https://toc.library.ethz.ch/objects/pdf03/e01_978-0-444-
63589-1_01.pdf  
[5]  V.  B.  L.G.  Austin,  "Experimental  methods  for  grinding 
studies in laboratory mills," Powder Technology, vol. 5, no. 
5,  pp.  261-266,  2020.  https://doi.org/10.1016/0032-
5910(72)80029-9  
[6]  B.  Wills  and  J.  Finch,  Wills’  Mineral  Processing 
Technology: An Introduction to the Practical Aspects of Ore 
Treatment and Mineral Recovery, Butterworth-Heinemann, 
2016.  https://shop.elsevier.com/books/wills-mineral-
processing-technology/wills/978-0-08-097053-0  
[7]  A. International, ASTM E11-17: Standard Specification for 
Woven  Wire  Test  Sieve  Cloth  and  Test  Sieves,  2017. 
https://standards.globalspec.com/std/3863958/astm-e11-17  
[8]  F.  Etzler  and  R.  Deanne,  "Particle  size  analysis:  A 
comparison  of  various  methods  ii,"  Particle  &  Particle 
system characterization, vol. 14, no. 6, pp. 278-282, 2004. 
https://doi.org/10.1002/ppsc.19970140604  
[9]  A. Freire and M. A. Lalbay Fuentes, "Implementation of a 
hammer  mill  for  the  wheat  flour  process,"  Technical 
University  of  Cotopaxi,  La  Maná,  2022. 
http://repositorio.utc.edu.ec/handle/27000/8473  
[10] F.  Lyu,  A.  van  del  Poel,  W.  Hendriks  and  M.  Thomas, 
"Particle  size  distribution  of  hammer-milled  maize  and 
soybean meal, its nutrient composition and in vitro digestion 
characteristics," Animal Feed Science and Technology, vol. 
281,  p.  115095,  2021. 
https://doi.org/10.1016/j.anifeedsci.2021.115095  
[11] R. Mugabi, Y. B. Byaruhanga, K. Eskridge and C. Welle, 
"Performance evaluation of a hammer mill during grinding 
of maize grains," AgricEngInt: CIGR, vol. 21, no. 2, p. 170, 
2019. 
https://cigrjournal.org/index.php/Ejounral/article/download
/5257/3011/23827?utm_source=chatgpt.com  
[12] P.  Pagán,  R.  Mollehuara-Canales,  M.  Sinche  and  J. 
Martínez,  Mechanical  Operations  of  Solid-Liquid 
Separation,  Spain:  Mc  Graw  Hill  Interamericana,  2023. 
https://books.google.com.ec/books?hl=es&lr=&id=UYftE
AAAQBAJ&oi=fnd&pg=PA2&dq=P.+Mart%C3%ADnez
-
Pag%C3%A1n,+Tecnolog%C3%ADa+Mineral%C3%Argi
ca,+Cartagena:+Universidad+T%C3%A9cnica+de+Cartag
ena,+2011.+&ots=XT4vyqxq_5&sig=apv8WuHNxMtMie
pueo-IAccQsUU&red  
[13] H. G. Merkus, Production, Handling and Characterization of 
Particulate  Materials,  The  Netherlands:  Springer  Cham, 
2015. https://doi.org/10.1007/978-3-319-20949-4  
[14] J. Fu, Z. Xue, Z. Chen and L. Ren, "Experimental study on 
specific  grinding  energy  and  particle  size  distribution  of 
maize  grain,  stover  and  cob.,"  International  Journal  of 
Agricultural and Biological Engineering, vol. 13, no. 4, p. 
135–142,  2020. 
https://doi.org/10.25165/j.ijabe.20201304.5327  
[15] G. Niu, T. Zhang, S. Cao, X. Zhang and L. Tao, "Effect of 
Corn Grinding Methods and  Particle Size on the Nutrient 
Digestibility of Chahua Chickens.,"  Animals, vol. 13,  no. 
14, p. 2364, 2023. https://doi.org/10.3390/ani13142364  
[16] P. Tomach, "The Influence of the Grinding Media Diameter 
on  Grinding  Efficiency  in  a  Vibratory  Ball  Mill,"  MDPI, 
vol.  17,  no.  12,  p.  2924,  2024. 
https://doi.org/10.3390/ma17122924  
[17] Y.  Catillo  Álvarez,  R.  Jiménez,  J.  Monteagudo,  B. 
Rodríguez and C. Patiño, "Mathematical Model to Improve 
Energy Efficiency in Hammer Mills and Its Use in the Feed 
Industry: Analysis and Validation in a Case Study in Cuba," 
Processes,  vol.  13,  no.  5,  p.  1523.,  2025. 
https://doi.org/10.3390/pr13051523  
[18] T.  Son,  H.  B.  Trinh,  S.  Kim,  B.  Dugarjav  and  J.  Lee, 
"Estimation  of  Energy  Consumption  for  Concentrate 
Process  of  Tungsten  Ore  towards  the  Integration  of 
Renewable Energy Sources in Mongolia," MInerals, vol. 13, 
no. 8, p. 1059, 2023. https://doi.org/10.3390/min13081059  
[19] P.  Martínez-Pagán,  M.  Sinche González  and  J.  Martínez, 
Mechanical  Operations  of  Solid-Liquid  Separation, 
Cartagena:  Mc  Graw  Hill,  2023. 
https://dialnet.unirioja.es/servlet/libro?codigo=990926   
[20] W.  Gutiérrez,  "Granulometric  Soil  Testing  Using  the 
Sieving  Method,"  Ciencia  Latina  Revista  Científica 
Multidisciplinar,  vol.  7,  no.  2,  pp.  6923-69-25,  2023. 
https://doi.org/10.37811/cl_rcm.v7i2.5834  
[21] A. Roca, "Standardization of the sieving method to establish 
acceptance and/or rejection criteria in raw materials for an 
extruded  line  in  the  food  industry,"  Universidad  de 
Antioquia,  Medellín,  Colombia,  2021. 
https://bibliotecadigital.udea.edu.co/server/api/core/bitstrea
ms/beaed1f9-a1d0-49d2-a007-1993bc2de107/content  
[22] A.  Terence,  Powder  Sampling  and  Particle  Size 
Determination,  Cartagena:  Universidad  Politécnica  de 
Cartagena,  2023. 
https://books.google.com.ec/books?hl=es&amp;lr=&amp;i
d=5NgqTf9L63kC&amp;oi=fnd&amp;pg=PP1&amp;dq=.
+Terence,+Powder+Sampling+and+Particle+Size+Determi
nation,+Cartagena:+Universidad+Polit%C3%A9cnica+de+
Cartagena.+(2020).,+2023.&amp; 
ots=1B0GcyOWcy&amp; 
sig=QkuduBccOmwYOrSmeJxjTPD1  
[23] P. Mort, "Analysis and graphical representation of particle 
size  distributions,"  Powder  Technology,,  vol.  420,  p. 
118100,  2023. 
https://doi.org/10.1016/j.powtec.2022.118100  
[24] G. Lyman, "A Statistical Theory for Sampling of Particulate 
Materials,"  Minerals,  vol.  13,  no.  7,  p.  905,  2023. 
https://doi.org/10.3390/min13070905  
[25] H.  Jung, Y.  Ju Lee and  W. B. Yoon, "Effect of Moisture 
Content on the Grinding Process and Powder Properties in 
Food:  A  Review,"  Processes,  vol.  6,  no.  6,  p.  69,  2018. 
https://doi.org/10.3390/pr6060069  
[26] M. Warechowska, "Some physical properties of cereal grain 
and  energy  consumption  of  grinding,"  Agricultural 
Engineering,  vol.  1,  no.  149,  pp.  239-2049,  2014. 
http://dx.medra.org/10.14654/ir.2014.149.025  
[27] H.  V.  ,.  J.  Dodds,  "Particle  shape  characterization  using 
morphological  descriptors,"  Particle&  Particle  systems 
characterization,  vol.  14,  no.  6,  pp.  272-277.  ,  2004. 
https://doi.org/10.1002/ppsc.19970140603.  

INQUIDE

Chemical Engineering & Development

Journal of Science and Engineering

Vol. 08 / Nº 01

e – ISSN: 3028-8533

ISSN – L: 3028-8533

Chemical Engineering & Development

University of Guayaquil | Faculty of Chemical Engineering

Guayaquil – Ecuador

https://revistas.ug.edu.ec/index.php/iqd

Email: inquide@ug.edu.ec

francisco.duquea@ug.edu.ec

Pag. 116

[28] X. Tian, X. Wang and Z. Wang, "Particle size distribution

control during wheat milling: nutritional quality and

functional basis of flour products—a comprehensive

review," International Journal of Food Science +

Technology, vol. 57, no. 12, p. 7556–7572, December 2022.

https://doi.org/10.1111/ijfs.16120

[29] N. Mohamad, I. Rodriguez-Donis, A. Cavaco-Soares and P.

Evon, "Bio-Refinery of Oilseeds: Oil Extraction, Secondary

Metabolites Separation towards Protein Meal

Valorisation—A Review," Processes, vol. 10, no. 5, p. 841,

2022. https://doi.org/10.3390/pr10050841

[30] S. Li, S. Ge, Z. Huang, Q. Wang and H. Zhao, , "Cryogenic

grinding technology for traditional Chinese herbal

medicine," Cryogenics, vol. 31, no. 2, pp. 136-137, 1991.

https://doi.org/10.1016/0011-2275(91)90260-4

[31] N. Yancey, C. T. Wright and T. L. Westover, "Optimizing

hammer mill performance through screen selection and

hammer design," Biofuels, vol. 4, no. 1, pp. 85-94, 2014.

https://doi.org/10.4155/bfs.12.77

[32] W. B. Rowe, "Towards High Productivity in Precision

Grinding," MPDI, vol. 3, no. 2, p. 24, 2018.

https://doi.org/10.3390/inventions3020024

[33] D. Wang, C. He, H. Tian, F. Liu and T. Zhang, "Parameter

optimization and experimental research on the hammer

mill," Imateh- Agricultural Enginerering, vol. 62, no. 3, pp.

342-349, 2020. https://doi.org/10.35633/inmateh-62-36

[34] F. Lyu, M. Thomas and W. H. Hendriks, "Particle size

distribution of hammer-milled maize and soybean meal, its

nutrient composition and in vitro digestion characteristics,"

Animal Feed Science and Technology, vol. 281, 2021.

https://doi.org/10.1016/j.anifeedsci.2021.115095

[35] R. K. Gupta and S. K. Das, "Fracture resistance of sunflower

seed and kernel to compressive loading," Journal of Food

Engineering, vol. 46, no. 1, pp. 1-8, 2000.

https://doi.org/10.1016/S0260-8774(00)00061-3

[36] Y. Haddad, F. Mabille, J. Abecassis and J. Benet,

"Rheological properties of wheat endosperm with a view on

grinding behaviour," Powder Technology, vol. 105, no. 1-3,

pp. 89-94, 1999. https://doi.org/10.1016/S0032-

5910(99)00122-9

[37] J. Ke, X. Wang, X. Gao and Y. Zhou, "Ball Milling

Improves Physicochemical, Functionality, and

Emulsification Characteristics of Insoluble Dietary Fiber

from Polygonatum sibiricum," MDPI, vol. 13, no. 15, p.

2323, 2024. https://doi.org/10.3390/foods13152323

[38] Y. Xie, C. Zhang and S. Mei, "Optimisation decision of

machining process parameters considering milling energy

consumption and specific cutting energy," Alexandria

Engineering Journal, vol. 128, pp. 786-795, 2025.

https://doi.org/10.1016/j.aej.2025.07.034

[39] L. Feltner, E. Korte, D. F. Bahr and P. Mort, "Particle size

and shape analyses for powder bed additive manufacturing,"

Particuology, vol. 101, pp. 33-42, 2025.

https://doi.org/10.1016/j.aej.2025.07.034

[40] Y. Liu, J. Wang, J. C. Barth, K. R. Welsch, V. McIntyre and

M. P. Wolcott, "Effects of multi-stage milling method on the

energy consumption of comminuting forest residuals,"

Industrial Crops and Products, vol. 145, 2020.

https://doi.org/10.1016/j.indcrop.2019.111955

[41] W. Kruszelnicka, Z. Chen and K. Ambrose, "Moisture-

Dependent Physical-Mechanical Properties of Maize, Rice,

and Soybeans as Related to Handling and Processing,"

Materials, vol. 15, no. 24, p. 8724, 2022.

https://doi.org/10.3390/ma15248729

[42] R. Politiek, M. Bruins, J. Keppler and M. Schutyser, "Effect

of oil content on pin-milling of soybean," Journal of Food

Engineering, vol. 334, p. 111149, 2022.

https://doi.org/10.1016/j.jfoodeng.2022.111149