Silica sand for water filtration in Morocco: D10, Cu and technical specifications

Key points for selecting filtration sand

Our analysis of Moroccan and international references leads to a clear conclusion: silica sand for water filtration must be selected as a technical component of the treatment process, not as ordinary bulk material.

A description such as “washed sand,” “fine sand” or “0.5–1 mm sand” is not sufficient on its own to predict how the medium will behave inside a filter.

To review a requirement correctly, we examine:

• the filter technology;
• the quality of the water to be treated;
• the filtration rate;
• the filter-bed depth;
• the complete particle-size distribution;
• D10 and D60;
• the uniformity coefficient Cu;
• the fine-particle content;
• the chemical and mineralogical composition;
• the underdrain system;
• the backwashing method;
• the required controls;
• the quantity, packaging and delivery destination.

These parameters influence water flow through the bed, head loss, cycle duration, backwashing quality and operating consistency.

Technical decision	Data to examine	Main impact
Define the media size	Particle-size curve, D10 and D60	Permeability, retention and head loss
Check grading consistency	Uniformity coefficient Cu	Size distribution and segregation risk
Control startup conditions	Fines and cleanliness	Initial turbidity, rinsing and clogging
Assess the nature of the material	Chemical and mineralogical analyses	Composition, stability and secondary constituents
Prepare the backwash process	Size, density, shape and bed depth	Bed expansion and media-loss risk
Secure the supply	Sample, analysis and batch traceability	Consistency between specification and delivered batch

Values published by ONEE, WHO, EPA and other institutions must always be interpreted within the process for which they were established. A specification used for slow sand filtration should not automatically be copied into a rapid-filtration or wastewater-treatment project.

How does a sand bed filter water?

A filter bed is a porous medium created by the arrangement of solid grains. Water flows through the voids between those grains.

Filtration is not simply a sieve that blocks everything larger than a fixed opening. Several mechanisms may act at the same time:

1. Local straining, when particles cannot pass through available pore openings.
2. Interception, when a particle following a streamline comes into contact with a grain.
3. Sedimentation within pores, when hydraulic conditions allow particles to settle.
4. Attachment to surfaces, influenced by the properties of the water, flocs and media.
5. Depth filtration, when particles are retained at different levels throughout the bed.
6. Biological activity, which is particularly important in slow sand filtration.

Performance therefore depends on the media and on the full treatment train:

• raw-water quality;
• coagulation and flocculation;
• clarification;
• suspended-solids loading;
• actual filtration rate;
• hydraulic distribution;
• bed depth;
• filter maturity;
• backwashing frequency and quality.

WHO notes that rapid filtration without suitable chemical pretreatment may behave mainly as a straining process and may not provide a sufficient microbiological barrier [4].

Filtration and disinfection perform different functions

Filtration may reduce turbidity, suspended matter and certain microorganisms depending on the process.

It does not automatically replace:

• disinfection;
• microbiological monitoring;
• chemical adjustment;
• treatment steps required by the water source;
• monitoring of the produced water.

We therefore consider the sand to be an integrated part of a treatment system, not an independent solution for producing drinking water.

Slow, rapid, dual-media and pressure filtration

Comparing particle-size ranges without identifying the filter technology can easily lead to an unsuitable specification.

Slow sand filtration

Slow filtration operates at a low hydraulic loading rate and relies in part on biological mechanisms developing near the top of the bed.

WHO documents indicate rates of approximately 0.1 to 0.3 m³/m²·h, equivalent to 0.1 to 0.3 m/h, for the slow sand filtration configurations discussed [2–3].

The process is characterized by:

• low filtration rates;
• a large required surface area;
• significant biological activity;
• specific maintenance practices;
• sensitivity to influent-water quality.

Open rapid filtration in the ONEE reference

The ONEE technical specification for drinking-water treatment describes, among other systems, an open, single-media, downward-flow filter.

For this configuration, the document specifies:

• homogeneous sand within the 0.5–1.5 mm interval;
• a uniformity coefficient of Cu ≤ 1.6;
• a sand-bed depth of 0.8 to 1.2 m;
• a filtration rate of ≤ 7 m/h;
• a rate not exceeding 9.5 m/h during the washing of one filter, under the conditions described;
• a maximum filtration cycle of 24 hours [1].

These values demonstrate the level of detail expected in a Moroccan project. They describe one specific configuration and should not be treated as a universal specification.

Sand–anthracite dual-media beds

The same reference presents a dual-media configuration using sand and anthracite.

It mentions:

• an effective size of approximately 1 mm;
• a filtration rate not exceeding 10 m³/m²·h;
• air and treated-water backwashing according to the planned sequence [1].

For a dual-media bed, selection must consider the relationships between:

• grain size;
• media density;
• depth of each layer;
• fluidization velocity;
• backwashing behavior;
• risk of layer mixing or inversion.

Pressure filters

The ONEE document also describes, in a section on brackish-water pretreatment, pressure filters containing sand with an effective size of approximately 0.8 mm and operating at a rate close to 10 m³/m²·h [1].

Pressure-filter projects require a detailed review of:

• the manufacturer’s datasheet;
• vessel geometry;
• filter surface area;
• operating pressure;
• underdrain and nozzle openings;
• backwash rate and sequence.

Intermittent wastewater sand filters

The EPA factsheet on intermittent sand filters gives, for this particular technology:

• an effective size of 0.25 to 0.75 mm;
• a uniformity coefficient below 4;
• bed depths generally between 18 and 36 inches [5].

A complementary EPA assessment also highlights effective size, uniformity coefficient and control of very fine fractions [7].

These criteria apply to intermittent filters receiving pretreated wastewater. They should not be transferred directly to rapid drinking-water filtration.

Documented hydraulic loading rates for three configurations

The difference in scale shows why the filter technology must be identified before selecting the media.

Slow sand filtration — WHO 0.3 m/h

Upper limit of the approximately 0.1 to 0.3 m/h range presented by WHO.

Open rapid filtration — ONEE 7 m/h

Maximum filtration rate indicated for the documented open-filter configuration.

Dual-media filtration — ONEE 10 m/h

Maximum filtration rate indicated for the cited sand–anthracite configuration.

Our analysis confirms that there is no universally “best filtration sand” independent of the process. Selection begins with the filter technology, hydraulic regime and treatment objective.

Understanding D10, D60 and the uniformity coefficient

A commercial range such as 0.5–1.0 mm identifies two limits, but it does not show how the material is distributed between them.

A sieve analysis produces a cumulative particle-size distribution curve from which several characteristic diameters can be calculated.

D10: effective size

D10 is the grain diameter below which 10% of the sample mass is finer.

Example:

• D10 = 0.50 mm;
• 10% of the mass is finer than 0.50 mm;
• 90% of the mass is coarser.

D10 is not necessarily the smallest grain present in the batch.

D60

D60 is the grain diameter below which 60% of the sample mass is finer.

It is mainly used with D10 to describe the width of the distribution.

Uniformity coefficient

The uniformity coefficient is generally calculated as:

Cu = D60 / D10

Example:

• D10 = 0.50 mm;
• D60 = 0.75 mm;
• Cu = 1.50.

A value close to 1 indicates a narrow distribution. A higher value indicates a wider distribution.

The ONEE reference specifies Cu ≤ 1.6 for the open filtration configuration described above [1]. EPA accepts a value below 4 for certain intermittent wastewater filters [5].

This difference is not a contradiction. The technologies, hydraulic loading rates and treatment objectives are different.

Why is the commercial range insufficient?

Two products both marketed as 0.5–1.0 mm may have:

• different D10 values;
• different D60 values;
• different Cu values;
• different fine-particle contents;
• different distributions around the median;
• different hydraulic behavior.

A complete particle-size curve provides far more useful information than a commercial designation alone.

Relationship between grain size and permeability

In classical models of flow through granular media, permeability varies partly with the square of a representative grain diameter when the other parameters remain constant.

The following chart presents a relative index calculated using:

Relative index = (d / 0.50)²

The purpose is to show the theoretical sensitivity of permeability to grain size. It does not replace hydraulic testing or a design calculation.

Theoretical sensitivity of permeability to grain size

Relative index calculated while keeping porosity, grain shape and the other characteristics of the medium constant.

Relative index based on a d² relationship

This calculation highlights an important point: a relatively small change in grain size may produce a significant hydraulic difference.

To compare two proposals, we recommend examining at least:

• the complete curve;
• D10;
• D50;
• D60;
• Cu;
• the fraction below the minimum sieve;
• the fraction above the maximum sieve;
• the test method;
• the batch number.

Particle size, head loss and filtration quality

Media selection generally seeks a balance between:

• particle-retention capacity;
• hydraulic flow;
• cycle duration;
• backwashing efficiency.

Media that is too fine

Depending on the process, an excessively fine fraction can:

• increase initial head loss;
• accelerate pressure-drop development;
• shorten the interval between backwashes;
• extend the initial rinsing period;
• increase the risk of fines migrating toward the underdrain;
• make the bed more sensitive to insufficient pretreatment.

Media that is too coarse

Media that is too coarse may:

• allow more particles to pass;
• lead to earlier turbidity breakthrough;
• require a different bed depth;
• be incompatible with underdrain openings;
• change the distribution of particle retention through the bed;
• require a different multi-layer arrangement.

Influence of pretreatment

Rapid-filter performance does not depend on the sand alone.

It is strongly influenced by:

• coagulant dose;
• pH;
• mixing quality;
• floc formation;
• clarification;
• influent turbidity;
• actual hydraulic loading.

Poor coagulation or filter overloading will not automatically be corrected by using finer sand.

Fines, cleanliness and media reception

Fine particles may come from:

• the deposit;
• crushing;
• drying;
• screening;
• attrition;
• transportation;
• loading operations;
• storage.

Their presence may cause:

• substantial turbidity during initial rinsing;
• rapid head-loss development;
• accumulation near the top of the bed;
• increased backwash duration;
• migration toward the underdrain.

Useful questions before ordering

We recommend checking:

1. Is the material washed?
2. Is it dried or supplied with variable moisture?
3. Which method is used to determine the fine fraction?
4. What percentage passes below the minimum sieve?
5. How are batches identified?
6. Does the sieve-analysis report correspond to the proposed batch?
7. How is the product protected during storage?
8. Which packaging method will be used?
9. Can a representative sample be reviewed?
10. Which controls are performed before shipment?

Receiving inspection

Depending on project criticality, receiving inspection may include:

• vehicle or packaging inspection;
• batch-number verification;
• inspection of bags or big bags;
• representative sampling;
• moisture measurement;
• confirmation sieve analysis;
• quantity verification;
• separated storage while awaiting approval.

A sample collected from only one location may not represent the full batch. The sampling plan should account for packaging and potential heterogeneity.

Mineralogical composition, chemistry and stability

The description “silica sand” is not an analytical result.

Depending on the project, characterization may include:

• X-ray fluorescence for major elements or oxides;
• X-ray diffraction for mineral phases;
• loss on ignition;
• solubility in a defined medium;
• organic-matter testing;
• examination of grain shape and surface;
• bulk-density measurement;
• resistance or attrition testing.

Requested parameters may include:

• SiO₂;
• Al₂O₃;
• Fe₂O₃;
• CaO;
• MgO;
• loss on ignition;
• organic matter;
• solubility;
• moisture;
• bulk density.

SiO₂ content is only one parameter

SiO₂ content may help document the nature of the material, but it does not guarantee:

• particle-size distribution;
• D10 or D60;
• uniformity coefficient;
• fine-particle content;
• compatibility with underdrain openings;
• head loss;
• backwash expansion;
• treated-water quality.

Conversely, a suitable particle-size curve does not replace the chemical or mineralogical controls required by the project.

French reference NF EN 12904

ANSES mentions silica-based sands and gravels complying with NF EN 12904 among mineral media used for water treatment [8].

For a Moroccan project involving French or European requirements, we examine:

• the cited standard;
• the applicable edition;
• additional requirements in the technical specification;
• project-specific sanitary requirements;
• requested test methods;
• required documentation.

A product standard does not remove the need for the designer to define the hydraulic and particle-size characteristics required for the installation.

What the ONEE reference contributes to Moroccan projects

The ONEE technical specification shows that a filtration project involves far more than supplying sand.

The documented open-filter configuration covers:

• the particle-size interval;
• Cu;
• bed depth;
• filtration rate;
• maximum cycle duration;
• clogging monitoring;
• filter-floor design;
• nozzle distribution;
• backwashing sequence;
• annual media loss [1].

Parameter in the reviewed reference	Requirement in the cited configuration	Purchasing consequence
Particle-size interval	0.5–1.5 mm	Request a curve demonstrating compliance with the specified envelope
Uniformity coefficient	Cu ≤ 1.6	Obtain D10 and D60
Sand-bed depth	0.8–1.2 m	Calculate the media volume for each filter
Filtration rate	≤ 7 m/h	Check consistency with the hydraulic load
Maximum cycle	24 h	Link the media to operating objectives
Annual sand loss	≤ 2%	Plan monitoring and replacement media
Backwashing	Air and water according to design	Check fluidization and media-loss risk

These values should be incorporated into a purchase only when the project actually adopts them. The project specification, design calculation and equipment datasheet remain the controlling documents.

Our role is to compare the client’s requirements with the parameters that are genuinely requested, without turning a documentary value into a general product promise.

Documentary contributions from France and Spain

Industrial proximity to France and Spain makes their technical references useful for Moroccan projects involving engineering firms, equipment manufacturers or European standards.

France

French documents may help review:

• product standards;
• sanitary requirements;
• test methods;
• public swimming-pool treatment;
• authorized or assessed filtration media;
• required technical documentation.

The ANSES swimming-pool report also highlights the importance of effective filtration and coagulation in the relevant systems [9].

Spain

The Spanish public project for La Granja de San Ildefonso describes a treatment line including:

• rapid mixing;
• flocculation;
• lamella clarification;
• open sand filtration;
• treated-water storage [10].

This arrangement confirms that filter performance must be assessed in relation to the upstream treatment steps.

How we use these references

During our review of a project, they may help us:

• clarify terminology;
• understand a cited standard;
• compare test methods;
• prepare multilingual documentation;
• interpret an equipment-supplier requirement;
• structure receiving documentation.

They do not replace Moroccan requirements or the project-specific technical specification.

Bed depth, support layers and underdrain systems

Depending on its design, a filter may include:

1. A main sand layer.
2. Anthracite or another upper medium.
3. Several graded gravel layers.
4. A false floor.
5. Nozzles.
6. Strainers.
7. An underdrain network.
8. An air-scour system.
9. A water-backwash system.

Purpose of support layers

Gravel support layers may:

• prevent sand migration;
• create a transition between the media and underdrain;
• protect strainers;
• distribute backwash water;
• stabilize the bed.

Their size must be compatible with:

• the sand size;
• underdrain openings;
• adjacent support layers;
• flow direction;
• backwash velocity.

Unsuitable sizing may cause:

• media loss;
• layer mixing;
• blockage;
• uneven backwashing;
• dead zones;
• reduced filtrate quality.

Volume calculation

The theoretical media volume is calculated as:

Volume = filtration surface area × bed depth

Methodological example:

• surface area: 25 m²;
• sand depth: 1 m;
• volume: 25 m³.

Conversion to mass requires the bulk density of the product in its delivery condition.

The calculation should also include:

• filter tolerances;
• settling;
• moisture;
• handling losses;
• initial reserve media;
• the maintenance strategy.

Backwashing and bed expansion

During filtration, retained particles gradually increase head loss.

Backwashing aims to:

• detach deposits;
• place the grains in motion;
• remove accumulated matter;
• restore hydraulic capacity.

The required rate depends on:

• grain size;
• density;
• grain shape;
• bed porosity;
• temperature;
• water viscosity;
• media depth;
• the presence of anthracite;
• the distribution system.

Effect of temperature

Water viscosity changes with temperature.

A constant backwash rate can therefore lead to:

• insufficient bed expansion;
• excessive bed expansion;
• media loss;
• incomplete cleaning.

The operating range should be assessed under the actual conditions of the installation.

Operating indicators

EPA guidance recommends examining the condition of the media, filter profile, hydraulic loading, underdrain system and backwashing practices when investigating filtration performance [6].

Useful indicators include:

• influent and effluent turbidity;
• differential pressure;
• cycle duration;
• backwash-water volume;
• backwash duration;
• observed bed expansion;
• media loss;
• turbidity after restart;
• differences between identical filters.

A deviation may indicate:

• poor coagulation;
• excessive loading;
• insufficient backwashing;
• damaged nozzles;
• bed segregation;
• media loss;
• uneven hydraulic distribution.

Laboratory controls to consider

The analytical program should match the project and the criticality of the installation.

Parameter	Possible data or method	Purpose
Particle-size curve	Sieve analysis using a defined method	Verify the complete distribution
D10	Reading or interpolation from the curve	Determine effective size
D60	Reading or interpolation from the curve	Calculate Cu
Cu	D60 / D10	Characterize the width of the distribution
Fines	Fraction below the specified sieve	Evaluate dust and very fine grains
Moisture	Drying and weighing	Determine actual dry mass
Bulk density	Volumetric measurement	Convert between volume and mass
Chemical composition	XRF or another suitable method	Document major constituents
Mineral phases	XRD	Identify quartz and other phases
Solubility	Project-defined method	Assess chemical stability
Organic matter	Appropriate method	Investigate certain contaminants
Attrition	Defined test	Evaluate potential generation of fines

Report traceability

A useful report should identify, as far as possible:

• the product;
• the fraction;
• the batch;
• the sampling date;
• the sample origin;
• the method;
• the laboratory;
• the units;
• the results;
• the person or organization validating the document.

A report not linked to the proposed batch does not provide the same level of control as a traceable result.

Building a usable technical specification

A well-prepared technical specification reduces ambiguity and makes quotations easier to compare.

Identify the application

Specify:

• drinking water;
• industrial water;
• process water;
• swimming pools;
• wastewater;
• membrane pretreatment;
• iron removal;
• manganese removal;
• safety filtration;
• another process.

Describe the filter

Indicate:

• open or closed;
• gravity or pressure;
• single-media or multimedia;
• surface area or diameter;
• available depth;
• flow rate;
• filtration rate;
• flow direction;
• underdrain design;
• backwashing method;
• media currently installed.

Define the grading

Specify as required:

• particle-size envelope;
• D10;
• D60;
• maximum Cu;
• maximum fine content;
• maximum coarse fraction;
• analytical method;
• tolerance between batches.

Define the other characteristics

Depending on the project:

• chemical composition;
• mineral phases;
• bulk density;
• moisture;
• solubility;
• loss on ignition;
• organic matter;
• attrition resistance;
• grain shape.

Define the documentation

The file may require:

• technical datasheet;
• certificate of analysis;
• particle-size report;
• sample;
• batch traceability;
• declaration of conformity;
• third-party laboratory report;
• standards documentation.

Define logistics

Specify:

• quantity;
• delivery tolerance;
• bags, big bags or bulk;
• unit weight;
• pallets;
• protection from moisture;
• destination;
• site access;
• unloading;
• schedule;
• possible split deliveries.

Cahier des charges

Ready-to-use request template

Complete the available information and attach your technical specification, equipment datasheet or laboratory results.

Copy the template

01

Projet

02

Client / bureau d’études

03

Application

04

Qualité de l’eau à traiter

05

Type de filtre

06

Marque / modèle

07

Surface du filtre

08

Hauteur du lit

09

Vitesse de filtration

10

Mode de lavage

11

Vitesse de lavage

12

Granulométrie recherchée

13

D10 recherché

14

D60 recherché

15

Coefficient d’uniformité maximum

16

Teneur maximale en fines

17

Composition recherchée

18

Autres essais

19

Quantité

20

Conditionnement

21

Destination

22

Date souhaitée

23

Documents à fournir

24

Observations

Ten common purchasing mistakes

1. Ordering only a nominal range

A range such as 0.5–1 mm does not define D10, D60, Cu or fine-particle content.

2. Automatically equating washed sand with qualified filter media

Washing alone does not demonstrate particle-size, chemical or hydraulic suitability.

3. Using a report from another batch

The raw material and preparation process may vary. The report should be linked to the proposed product.

4. Copying a specification found online

A value suitable for slow filtration, swimming pools or wastewater may be unsuitable for rapid drinking-water filtration.

5. Ignoring strainers and underdrains

Incompatibility can cause sand loss, blockage or uneven backwashing.

6. Ignoring backwash-water temperature

Bed expansion changes with water viscosity.

7. Converting volume using the true density of quartz

Mineral density and the bulk density of a sand bed are not interchangeable.

8. Forgetting moisture

Variable moisture affects the actual dry mass delivered.

9. Mixing batches without control

Different particle-size distributions may alter bed behavior.

10. Using the media to compensate for a process problem

Poor coagulation, hydraulic overloading or inadequate backwashing require a full process diagnosis.

Selection by application

Application	Main review priorities	Information to provide
Drinking water — rapid filter	D10, Cu, bed depth, turbidity and backwash	Specification, filtration rate, filter type and requested analyses
Pressure filter	Strainers, pressure, flow and backwash	Brand, model, diameter and manufacturer datasheet
Dual-media bed	Size, density and expansion of both media	Layer composition and backwashing sequence
Public swimming pool	Filtration rate, coagulation, turbidity and operation	Flow rate, filter type and applicable requirements
Process water	Final water quality and equipment protection	Contaminants, flow and downstream treatment steps
Wastewater	Pretreatment and hydraulic loading	Effluent quality and dosing method
Membrane pretreatment	Turbidity, pressure and membrane protection	Membrane type and water-quality objectives
Iron or manganese removal	Oxidation, pH and intended media	Fe/Mn analyses and process description

Moroccan research and sand characterization

The Moroccan studies reviewed reinforce a central principle: the technical value of sand must be established through characterization.

The study conducted on sand quarries in the Meknes region combined:

• geotechnical analyses;
• mineralogical analyses;
• geochemical analyses [12].

It demonstrates the importance of investigating the nature of the material before industrial use.

Another study used natural Moroccan sand to produce a ceramic microfiltration support [11].

That application differs from a conventional granular filter bed:

• the sand is processed;
• the material is shaped;
• it undergoes thermal treatment;
• filtration depends on the pores of a membrane.

This distinction matters. We do not use the publication to claim that raw sand is automatically suitable for a conventional filter. We retain the conclusion that Moroccan mineral resources can offer technical value when they are properly characterized and processed for a defined application.

Our approach follows the same logic:

1. understand the process;
2. identify the critical parameters;
3. review the available characteristics;
4. define the necessary analyses;
5. organize supply and documentation.

Procedure before an industrial order

Step 1 — Gather the available data

Collect:

• drawings;
• equipment datasheets;
• design calculations;
• operating history;
• existing media information;
• actual flow rate;
• backwashing settings;
• observed problems.

Step 2 — Define the objective

Identify:

• influent-water quality;
• expected treated-water quality;
• acceptable head loss;
• target cycle duration;
• downstream-process constraints;
• maintenance frequency.

Step 3 — Formalize media characteristics

Define:

• grading envelope;
• D10;
• D60;
• Cu;
• fines;
• composition;
• physical properties;
• documentation.

Step 4 — Review the proposal

Compare the available data point by point with the technical specification.

Every deviation should be:

• identified;
• explained;
• evaluated;
• submitted to the party responsible for technical approval.

Step 5 — Validate a sample

Depending on project criticality:

• sample;
• analyze;
• compare;
• document;
• retain a reference sample.

Step 6 — Plan delivery

Prepare:

• packaging;
• access;
• unloading;
• storage;
• protection;
• filling sequence;
• rinsing-water availability;
• packaging-waste management.

Step 7 — Control commissioning

After filling:

• verify the layers;
• level the bed;
• carry out rinsing;
• observe the wash water;
• monitor media loss;
• return the filter to service progressively;
• record initial operating values.

Conclusion: specify filter media, not simply sand

Our study shows that selecting silica sand for water filtration in Morocco requires four levels of analysis.

The process

• filter technology;
• water quality;
• pretreatment;
• filtration rate;
• backwashing;
• expected performance.

The media

• particle-size distribution;
• D10;
• D60;
• Cu;
• fines;
• grain shape;
• composition;
• stability.

Quality control

• sampling method;
• analyses;
• traceability;
• receiving inspection;
• batch monitoring.

Supply

• quantity;
• moisture;
• packaging;
• transport;
• storage;
• schedule.

ONEE references demonstrate the level of precision that Moroccan projects may require. Documents from WHO, EPA, ANSES and European institutions also show that values vary significantly depending on the technology.

A useful request should therefore not be limited to:

“We need sand for a filter.”

It should state:

“Here is our application, equipment, grading, D10, D60, Cu, requested tests, volume, packaging and destination.”

Based on this information, we can review the requirement clearly, compare the available characteristics with the project specification and support the client in preparing the supply.