Soil Classification Systems
Soil Classification Systems
Soil Classification Systems
Classification Systems
Soil Classification is the arrangement of soils into groups with similar physical, mechanical and
behavioral properties. Classification of soils allows for the prediction of a soils behavior under
loading, unloading or other conditions. Careful determination and communication of these
engineering properties is important for all design and construction purposes. (See ENG
PROPS=P .pdf for an example of a clear communication of soil classes and their relative
properties)
Several systems of classification have been developed and are still in practice today (see page
21 of Atkins text, note the variation in particle boundaries). Each has a specific intention or
purpose for soil classification. We will focus our attention to the Massachusetts Institute of
Technology (MIT) soil texture classification and Unified Soils Classification Systems (USCS).
1) Massachusetts Institute of Technology (MIT) soil texture classification:
Textural Classification is a method of classification made on the basis of soil texture (ie
nominal particle size). The ASTM and MIT systems are textural classifications. MIT has
become a standard method on logging soils during field activities. The MIT system is
based on the following naming convention.
MIT >50% use as a noun in CAPITALS. (i.e GRAVEL, SAND, SILT or CLAY)
50 – 35 % : and ( and gravel, and sand, etc)
35 – 15 % : adjective (gravely, sandy, silty, clayey)
15 – 5 % : “with some”
< 5 % : “trace of”
Example: % Boulder = 3%
% Gravel = 25.2 %
% Sand = 65.8 %
% Silt = 6.0 %
% Clay = 0.0 %
Answer: Gravely SAND with some silt and trace boulders
2) Unified Soils Classification System (USCS):
The Unified System is published as Standard D2487 by ASTM (The American Society for
Testing and Materials). This system is comprised of a series of two letter designations
which indicate the soil type, and also the gradation and plasticity.
For more details refer to;
1. “M.T.O.” SOIL CLASSIFICATION PDF.
2. Pages 24 to 28 of the Atkins text
3. and/or ASTM D2487.
Soil Classification – Laboratory Tests
Classification of soils implies a degree of assessment of the soils characteristics. This
assessment can be broadly categorized into three domains.
1. Visual/Manual techniques (eg field tests in Lab1 for the determination of silt and clay
content)
2. Laboratory Testing (discussed in more detail below)
3. In‐situ (in place) testing.(discussed in more detail in our Field Investigations Unit)
Soil is commonly classified according to the following two characteristic properties; Grain size
and Cohesion (Plasticity).
1. Grain size – The determination of the particle size distribution of fine‐grained soils (silt,
clay) and coarse‐grained soils (sand, gravel,), can accomplished through a Mechanical
(Sieve) and/or Sedimentation (Hydrometer) Analysis. A third option occasionally used
is the Wash Sieve.
o Sieve Analysis (Lab #3) – performed on coarse‐grained soils (sands and gravels)
If large quantities of cohesive grains are present, a wet/wash sieve may
be performed to determine the % fines, not distribution. This is
accomplished by washing a small portion of coarse material is a deep
#200 (0.075mm) sieve.
o Hydrometer (Lab #4) – performed on fine‐grained soils (silts and clays)
If significant quantities of both fine and coarse grains are present then
both sieve and hydrometer tests must be performed.
2. Cohesion (Plasticity) – Of the many different types of behavior a soil may demonstrate
under varying conditions, plasticity is the most recognizable behavior exhibited by a
fine‐grained soil under a shear stress.
o Atterberg Limits Tests (Lab #5 ) – performed on fine‐grained soils (silts and clays)
Grain Size
As discussed earlier soil formation may lead to the particles of various sizes with similar mineral
structure to the parent rock or new chemical arrangements as in clay mineral. Particle shapes
may range from angular to smooth and are a function of the soils age as well as the degree and
method of transport and deposition. This transportation and deposition may lead to a range of
particles sizes from narrow ranges where all particles are o similar sizes to a wide range of
particle sizes where boulders, cobbles, gravels, sands, silts and clays are mixed together.
Sieve Analysis (Lab #3): A sieve analysis begins with the mechanical separation of a coarse‐
grained sample in a stack of sieves. The sieves are stacked such that the sieves with the largest
apertures are at the top and the smallest apertures are at the bottom resting on a pan that
catches the remaining fines material. The result of a sieve analysis is the distribution of masses
in the separate sieves. The total mass of a sample is distributed among the sieves. The mass
distribution plotted to produce a Particle Size Distribution (PSD) curve aka Grain Size Curve
which is a semi log plot of grain size (ie nominal particle diameter) on the horizontal vs. %
passing/finer on the vertical. Refer to the grain size curve paper and PSD sample curves A, B,
and C as an illustration.
It is important to note that some of the variation in the mass distribution will result from a
variety of errors that may arise in the sieving process, they include: malformed sieves, torn or
warped screens, excessive sieving, insufficient sample preparation (moist sample), insufficient
sample size, excessive sample size, etc. The sample selection and preparation process must be
done with care if the exercise is to yield information that is representative of the whole. It is
often the case that points at either extreme of the curve are speculative because very large and
extremely small particles cannot be easily sieved.
It is also important to note that a soil mass resting on a particular sieve size, (eg the 0.6 mm
sieve) is not entirely a collection of soil particles, each with a particle diameter greater than 0.6
mm but rather a collection of soil particles with at least one dimension that is smaller than the
sieve directly above the sieve on which it now rests. Hence the mass and particle size is not
directly measured but estimated and therefore the points are estimates.
Soil Texture/Gradation ‐ the proportions by mass of a soil or fragmented rock
distributed in specified particle‐size ranges.
The shape of the grain size distribution curve along with field notes & observations and
research can yield information about the origin and placement of a material. The curve
placement and shape will also yield important information about the relative engineering
properties of a given soil.
Placement:
Coarse‐grained and fine‐grained soils can be distinguished by referring to the particle
diameter at the 50% passing mark on the curve. Coarse‐grained soils contain more than 50% by
weight of material that is larger than 0.075mm. Conversely, fine‐grained soils contain more
than 50% by weight of material that is smaller than 0.075mm.
A soil with a variety of particle sizes will often have small particles filling in the void
space between large particles in its matrix. Thus engineering properties such as the
permeability of a soil are related to the smallest particles within the sample. This yields
another distinction of a soil referred to as the Effective Diameter. The effective diameter is the
particle diameter at which 10% of the sample is smaller (ie D10).
Shape: broadly defined by the following three categories;
Well graded soils: have a large range of particle diameters
Poorly Graded Soil which can be further defined as;
o Uniform soil (aka well sorted): narrow range of particle diameters
o Gap graded soil: lacking particles of a range of diameters.
This qualitative description is determined by analyzing specific points on the curve, points
which provide reasonably accurate estimates of the soils breadth and completeness of
gradation. These points are standard percent passing values 60% 30% and 10% and the
associated particle diameter, expressed as D60, D30 and D10.
Uniformity Coefficient (CU). This value gives some indication of the shape of the curve and the
range of particle sizes that a soil contains, especially in the more important fine part of the soil.
The formula is:
ܦ
ܥ ൌ
ܦଵ
Coefficient of Curvature (CC). This is another measurement of the shape of the curve. It is
calculated as follows:
ሺܦଷ ሻଶ
ܥ ൌ
ܦଵ ܦݔ
Practice: Refer to the 2 figures (6 curves) above, identify the D10, D30 and D60 values,
calculate the coefficient of uniformity and coefficient of curvature and determine the
shape of each curve.
Class example: Using the information below, complete the following:
1. Calculate the % passing and plot the GSC.
2. Identify the % gravels, sands and fines
3. Classify the soil using the MIT system.
4. Uniformity coefficient (Cu), Cu = D60/D10.
5. Coefficient of curvature (Cc), Cc = (D30)2/(D60xD10)
6. Classify the soil using USCS.
Mass of Container (g): Mass of cont. & dry soil (g): Original sample mass (g):
9.5
4.75
2.36
1.18
0.6
0.3
0.15
0.075
Pan
Hydrometer (Lab #4)
Stokes Law is the application of the three principle forces, gravity, drag and buoyancy which act
on a spherical object falling at a terminal velocity in a viscous fluid. Stokes Law in its general
form may be applied to numerous viscous flow applications, in soil mechanics, it is used to
determine particle sizes because particles in a suspension settle out at a rate that varies with
their size.
The hydrometer test is used to determine the PSD of the fine‐grained fraction of a soil. Larger
particles settle out of suspension before smaller ones due to their higher terminal velocity. The
individual particle’s mass, diameter and terminal velocity are not directly measured. Rather, a
hydrometer is used to measure the density of the water–soil suspension at various times as the
grains settle. The rate at which a hydrometer bulb drops while suspended in a solution of
water, soap and soil can be correlated to suspended particle size. Using this law, we can
calculate the size of a particle that has settled a known distance in the suspension at any time
from the beginning of sedimentation. The size of the particle that has settled to the center of
the hydrometer bulb can be calculated, and the density of the solution indicates the percentage
of the sample still in the suspension. Using this data, a grain‐size distribution curve can be
established.
The following form of the Stokes Equation is derived assuming all particles have the same
relative density (2.6), are spherical, smooth and do not interact with each other, moreover, that
the solution is homogenous and that flow is laminar
ܮ
ܦൌ ܭඨ
ܶ
Where:
D is the particle diameter
K is a coefficient simplified from the assumed fluid and particle relative densities, and
temperature.
L is the length (cm) the bulb has dropped in time T (mins)
The percent of the sample still in suspension at the time of the hydrometer reading is given by:
ܴߙ
ܲൌ
ܯௌ
Where:
P is the percent smaller (ie still in suspension)
R is the corrected hydrometer reading. The hydrometer reading must be corrected to account
for changes in fluid density due to dissolved minerals, temperature changes, the addition of a
deflocculating agent (soap) and others.
α is a correction coefficient for the relative density of the particles.
MS is the original sample mass.