How to carry out Marshall mix design for Hot-Mix Asphalt (HMA) for flexible pavement

Hot mix asphalt is composed of asphalt and aggregates, heated and blended into a uniform mixture in precise proportions. When a sample of paving mixture is prepared in the laboratory, it can be analysed to determine its probable performance in a pavement structure. The analysis focuses on four (4) characteristics of the mixture and the influence of those characteristics are likely to have on the mix behaviour. The 4 characteristics are:

Four Characteristics of Hot Mix Asphalt

1. Mix Density
2. Air Voids
3. Voids in Mineral Aggregate (VMA) and
4. Asphalt Content

Density: Density of the compacted mix is its unit weight (the weight of specific volume of the mix). It is important because high density of the finished pavement is essential for lasting pavement performance. The density determined in the laboratory becomes the standard by which the density of the finished pavement is determined to be adequate or inadequate. This is because in-situ compaction rarely can achieve the densities achieved by standard laboratory compaction methods. To overcome this shortcoming, specifications usually require pavement density to be a percentage of laboratory density.

Air Voids: These are small air spaces or pockets of air that occur between the coated aggregate particles in the final compacted mix. A certain percentage of air voids is necessary in all dense-graded highway mixes to allow for some additional pavement compaction under traffic and to provide spaces into which small amounts of asphalt can flow during this subsequent compaction. The allowable percentage of air voids in laboratory specimen is between 3% and 5% for surface courses and base courses respectively depending on the specific design. The durability of an asphalt pavement is a function of the air-void content.

Density and air voids are directly related. The higher the density, the lower the percentage air voids in the mix, and vice versa. Job specifications usually require pavement density that allows as low an air void content as is practical, preferably less than 8%.

Voids in Mineral Aggregate (VMA): This is also important because the thicker the asphalt film on the aggregate particles, the more durable is the mix.

Asphalt Content: The proportion of asphalt in the mixture is critical and must be accurately determined in the laboratory and then precisely controlled in the field. The optimum asphalt content of the mix is highly dependent on the aggregate characteristics such as gradation and absorptiveness. Aggregate gradation is directly related to optimum asphalt content. The finer the mix gradation, the larger the total surface area of the aggregate and the greater the amount of asphalt required to uniformly coat the particles and vice versa. The absorptiveness (ability to absorb asphalt) of the aggregate used in the mix is critical in determining the optimum asphalt content, because enough asphalt must be added to the mix to allow for absorption and still coat the particles with adequate film. The absorptiveness of an aggregate is generally known for established aggregate sources, but requires careful testing where new aggregate sources are being used.

Mix Design of Flexible Pavement

Many methods are available for the mix design of asphalt for flexible pavement construction. These methods include but to limited to: Marshall method, Hveem method, Superpave method etc. Among these methods, the Marshall method is still very popular and it is the method commonly used in Nigeria.

The prime objective of the design is to determine the optimum binder/bitumen content for the asphalt. It is also expected that the design should possess the following properties: stability, durability, impermeability, workability, flexibility, fatigue resistance and skid resistance. The design of asphalt mix is a very critical process in flexible pavement construction because poor design can lead to porous asphalt which is liable to early failure under load. Even though asphalt is not the structural component of flexible pavement, the protection it gives other layers made predominantly of earth against water enables those layers to perform their function optimally. Thus, when the asphalt fails, other layers would soon fail as well. Presented below are the causes of undesirable properties in asphalt:

Causes of Low Stability

1. Excess asphalt in mix
2. Excess medium-sized sand in mix
3. Rounded aggregate, little or no crushed surfaces

Causes of Poor Durability

1. Low asphalt content
2. High void content through design or lack of compaction
3. Water susceptible (hydrophilic) aggregate in mixture

Causes of too Permeable Mix

1. Low asphalt content
2. High voids content in design mix
3. Inadequate compaction

Causes of Poor Workability

1. Large maximum-sized particle
2. Excessive coarse-aggregate
3. Too low a mix temperature
4. Too much medium-sized sand
5. Low mineral filler content
6. High mineral filler content

Causes of Poor Fatigue Resistance

1. Low asphalt content
2. High design voids
3. Lack of compaction
4. Inadequate pavement thickness

Causes of Poor Skid Resistance

1. Excess asphalt
2. Poorly textured or graded aggregate
3. Polishing aggregate in mixture

Figures 1 shows the Marshall test set-up. The Marshall method uses standard test specimens, 63.5 mm high and 102 mm in diameter.  A series of specimens each containing the same aggregate blend but varying in asphalt content, is prepared using specific procedure to heat, mix and compact the asphalt aggregate mixtures. The main aim being to determine the optimum asphalt content for particular blend of aggregate while the objectives are to obtain mix having:

1. Sufficient asphalt to ensure durable pavement
2. Adequate mix stability to satisfy the demands of traffic without distortion or displacement
3. Voids content high enough to allow for slight amount of additional compaction under traffic loading without flushing, bleeding, and loss of stability, yet low enough to keep off harmful air and moisture.
4. Sufficient workability to permit efficient placement of the mix without segregation.

The different aggregates and asphalt used have different characteristics and these characteristics have direct impact on the nature of the pavement itself. The first step in the design method is to determine what qualities (stability, durability, workability, skid resistance etc) that the paving mixture must have and to select a type of aggregate and a compatible type of asphalt that will combine to produce those qualities. Once this is done, the test preparations can begin.

The first step in test preparation is to gather samples of the asphalt and the aggregate that will be used in the actual paving mixture. It is very important that the asphalt samples and aggregate samples used should have similar characteristics as those to be used in the final hot-mix asphalt in the field. This is very important because past premature failure of pavement have been linked to variances between materials tested in the laboratory and the materials actually used in the field.

Figure 1; Marshall test set up

Procedure for the Design of Hot Mix Asphalt

The procedure for the design is itemized as follows:

• Heat bitumen of 3.5% or 4% or 5% by weight of the mineral aggregates to 121 – 125^oC temperature.
• Heat about 1200 g of aggregates and filler to a temperature of about 175 – 190^o
• Thoroughly mix the heated aggregate and bitumen at a temperature of 154 – 160^o
• Place the mix in pre-heated mould and compact by a rammer with 50 blows on either side at a temperature of 138 – 149^oC to a compacted thickness of 63.5 ± – 3mm (see Figure 1)
• Vary the bitumen content by +0.5% and repeat the first four steps above.
• Repeat the entire trial for about 5 – 7 times.

Calculation

The properties sought in the mix for the design are:

Theoretical or apparent specific gravity, G_t: This is the specific gravity without considering voids in the mix.

Bulk or actual specific gravity, G_m: This is the specific gravity considering air voids.

Percent air voids, V_v: This is the percent of air voids in relation to the volume of specimen.

Percent volume of bitumen, V_b: This is the percent of volume of bitumen in relation to the volume of specimen.

Percent void in mixed aggregate, VMA: this is the volume of voids in the aggregate.

Percent voids filled with bitumen, VFB: this is the volume of voids that is filled with bitumen in relation to the volume of specimen.

These properties are clearly described by the phase diagram shown below (Figure 2).

Figure 2; Phase diagram for asphalt

The properties are calculated by the following formulas

G_t = (W₁ + W₂ + W₃ + W_b)/((W₁/G₁) + (W₂/G₂) + (W₃/G₃) + (W_b/G_b))

Where,

W₁ = weight of coarse aggregate in the total mix

W₂ = weight of fine aggregate in the total mix

W₃ = weight of filler in the total mix

W_b = weight of bitumen in the total mix

G₁ = apparent specific gravity of the coarse aggregate

G₂ = apparent specific gravity of the fine aggregate

G₃ = apparent specific gravity of the filler

G_b = apparent specific gravity of the bitumen

Bulk specific gravity, G_m = W_m/ (W_m – W_w)

Where

W_m = weight of mix in air

W_w = weight of mix in water

N.B: W_m – W_w = Volume of mix

Percent air voids, V_v = ((G_t – G_m) x 100) / G_t

Where the terms are same as defined previously

Percent voids of bitumen, V_b = (W_b/G_b) / ((W₁ + W₂ + W₃ + W_b)/G_m)

Where the terms are same as defined previously

Voids in mineral aggregate, VMA = V_v + V_b

Where the terms are same as defined previously

Voids filled with bitumen, VFB = (V_b x 100) / VMA

Where the terms are same as defined previously

How to Determine Marshall Stability and Flow

After the design, Marshall stability and flow tests/density-voids analysis are used to measure the performance of the design. Marshall stability of a test specimen is the maximum load required to produce failure when the specimen is pre-heated to a prescribed temperature placed in a special test and the load applied at a constant strain. The stability portion of the test measures the maximum load supported by the test specimen at a loading rate of 50.8 mm/minute. Load is applied to the specimen till failure and the maximum load is designated as stability. While the stability test is in progress, dial gauge is used to measure the vertical deformation of the specimen. The deformation at the failure point expressed in units of 0.25 mm is called Marshall flow value of the specimen.

Stability Correction

Due to human errors, the thickness of the specimen may vary from the standard specification of 63.5 mm. There is need to do correction to values that should have been obtained had the specimen be exactly 63.5 mm. The correction factors are presented in Table 1.

Table 1: Marshall stability value correction factors

Table 2: Marshall mix design specification limits

Example

The table below shows results from design carried out for wearing course for a location in southern part of Nigeria. The Marshall stability was done according to ASTM-D1559. The specified limits used in the mix design are based on Federal Ministry of Work of the government of the Federal Republic of Nigeria specification (Revised 1997). The aggregates were tested based on the method described in B.S. 812.

Table 3; Summary results from Laboratory tests and calculations

Table 4; Summary results used to plot graphs 

Graphs from Table 4

The optimum binder content is the average value of the binder content corresponding to:

1. Maximum stability = 5.8 %
2. Maximum bulk specific gravity = 6 %
3. Median of the design limits of percent air voids (3.10%) = 6 %

Optimum binder content = (5.8+6+6)/3 = 18/3 = 5.93 %

Thanks!

References

Matthew, T.V. and Rao, K.V.K (2007): Introduction to Transportation Engineering. NPTEL