How to cite this paper : J.

Davidovits, (2018), Why Alkali-Activated Materials (AAM) are Not

Geopolymers, Technical Paper #25, Geopolymer Institute Library, www.geopolymer.org ,
DOI: 10.13140/RG.2.2.34337.25441

Why Alkali-Activated Materials (AAM) are NOT Geopolymers ?

Prof. Dr. Joseph Davidovits, Geopolymer Institute
(November 2018, script of the Video series available
at the Geopolymer Institute, https://www.geopolymer.org/faq/alkali-activated-materials-geopolymers/
and on YouTube).

Many scientists and civil engineers are mistaking alkali activation for geopolymers, fueling
confusion, using them as synonyms without understanding what they really are.
To sum up: Alkali-Activated Materials (AAM) are NOT Polymers, so they cannot be called Geo-
Polymers. Geopolymers are NOT a subset of AAM because they are not a calcium hydrate
alternative (no NASH, no KASH). They belong to two very different and separate chemistry
systems (a hydrate/precipitate that is a monomer or a dimer versus a true polymer). Those who
claim that both terms are synonyms are promoting a misleading scientific belief. Learn why by
watching these four videos.
In my four keynotes at the Geopolymer Camp (2014-2017), I explained why Alkali-
Activated-Materials are not Geopolymers, or why alkali-activation is not geopolymerization.
We have selected all the sequences that had been dedicated to this issue in the
GPCamp-2014, 2015, 2016 and 2017 keynotes. These videos are titled: Why Alkali-
Activated Materials are NOT Geopolymers. You will finally understand why there are two
different systems.

Part 1 (2014): AAM are not geopolymers, two different chemistry

I explain the main differences between AAC
(Alkali-Activated Concrete), AAS (Alkali-Activated
Slag), AAF (Alkali-Activated Fly Ash) and Slag-
Metkaolin-based Geopolymer cement, in terms of
chemistry, molecular structure, long-term
durability. In a second part, on hand of the
industrialization of Slag/fly ash-based geopolymer
cement/concrete implemented by the company
Wagners, Australia, I focus on the results
provided by the carbonation testing data obtained
for ordinary Portland cement, AAS and EFC
(Slag/fly ash-based geopolymer). The tests were
carried out at the Royal Melbourne Institute of
Technology RMIT in Australia. Geopolymer
behaves like regular Portland cement, whereas
AAS gets very bad carbonation results.

The slag is made out of a glassy substance that

is called “melilite’ which is a solid solution of
gehlinite (Ca-silico-aluminate) and akermanite
(Mg-silicate) (see in Fig. 1). When these two
minerals are subjected to alkalination (not
activation), they depolymerize under the action of
the alkali.

Figure 1: first step of slag alkalination.

The gehlinite (on the left) is transformed into (K,

Ca)-ortho-sialate-hydrate, a small molecule and
precipitation of aluminum hydroxide Al(OH)3.
Ackermanite (on the right) undergoes
depolymerization into the small molecule Ca-di-
silicate-hydrate that is the CSH of the cement
chemistry, plus the precipitation of magnesium
hydroxide Mg(OH)2.
 -2-
The two small molecules will interact (Fig. 2) and form a more complicated structure (on the left), a cycle
involving 3 Si, 1 Al, 1 K and 3 Ca, (K, Ca)-cyclo-
ortho (sialate-disiloxo), and also CSH Ca-di-
siloxonate-hydrate, plus precipitation of Al(OH)3,
Mg(OH)2 and Ca(OH)2.
The problem resulting from this reaction is
the following one: the alkali cation K+ is outside
of the structure of the cycle; same case for Na+.
This simply means that the cations will migrate
very fast in contact with water. This will generate
leachates and bad long-term properties. This is
generally what people are getting with AAS. This
is not geopolymer and not the end term of
geopolymerization.

Figue 2: alkali-activated slag unstable structure.

Therefore, all people who are claiming that alkali-

activation of slag is similar to geopolymerization
are totally wrong and essentially in a lot of
scientific papers. They are only doing this first
step of geopolymerization.

To continue geopolymerization into genuine

geopolymer cement, we must add a networking
element that will interact with the free cations (K+,
Na+). NMR spectroscopy (Fig. 3) shows the 29Si
resonance for AAS, with SiQ2 species. Adding the
networking molecule of metakaolin MK-750 we
obtain a Si(Q3) branched molecule that is bigger
and more stable, and we end up with a Si(Q4)
stable specie, a three-dimensional networking.

Figure 3: 29Si NMR spectrum of slag/MK-based

geopolymer, evolution from alkali-activated-slag
towards geopolymer cement.

The mechanism is summarized in Fig. 4. AAS

gets high strength, for example 90 MPa
compressive strength, vs “only” 60/70 MPa for
slag/MK-based geopolymer. But AAS generates
leachates, and is not long-term stable with poor physico-chemical properties because, in the chemical
structure of its hydroxylated oligomer, the Na+ or K+ cations are located on the edge of the cyclic
oligomer (K, Ca)—cyclo-ortho (sialate-disiloxo) with Si(Q2). They can easily migrate in contact with water
and be leached out. This is not the case for geopolymer cement with its three-dimensional networking.

From alkali-activated slag to stable 3D

geopolymer network resulting from the addition of
MK-750. From left to right, structural changes
from unstable cyclic oligomer Si(Q2) obtained by
alkali activation of blast furnace slag, through
ribbon bidimensional polymers Si(Q3) towards
stable reticulated 3D geopolymers with Si(Q4).

Figure 4: Hardening results from the

polycondensation of the ribbon bidimensional
polymers Ca—poly (sialate) Si:Al=1, and (Na, K)—
poly (sialate- disiloxo) Si:Al=3, with Si(Q3) towards
more stable reticulated 3D geopolymers with
Si(Q4).

This mechanism provides long-term stability

+ +
because the Na or K cations are fixed and trapped inside the frameworks of the types Ca–anorthite, Na
—albite or K–orthoclase (as in natural feldspar or plagioclase). In fact, by varying the ratio slag / MK-750,
 -3-
end-users can choose between high compressive strength and low stability (danger for alkali-activated
slag), or optimal strength with long-term durability
and corrosion resistance for geopolymer
cements.

Fly ash plays the same role as metakaolin. This

is exemplified in Fig. 5 which compares the
carbonation tests of slag-based geopolymer
cement (slag/fly ash EFC from Wagners) with
OPC plain, OPC with various slag GGBS content
and AAS alkali-activated-slag.

Figure 5: accelerated carbonation tests: EFC

geopolymer concrete gets 1.94 mm to 2.05 mm
CO2 penetration compared with 4.56 mm for AAS
(RMIT testing).

The carbonation tests for geopolymer concrete

and slag/blended Portland cement are similar.

Conclusion: AAM is not geopolymer.

Part 2 (2015): Clarifying statement

about all the false ideas and
assertions
I made a clarifying statement about all the false
ideas and assertions written by several alkali
activated materials scientists (incorrectly using
the word “geopolymer” for marketing purpose in
place of AAM) and blindly imitated by others. I
explain why it is a true polymer with a well-known
and understood chemistry (as opposed to those
who claim it is a “gel” of unknown character),
mentions the historicity and discovery of
geopolymer chemistry, the real contributions of
Glukhovsky and what he really wrote about
geopolymers. I develop the range of actual
industrial applications that goes far beyond
cement made out of wastes.
Got to: https://www.geopolymer.org/faq/alkali-activated-materials-geopolymers/ and on YouTube.

In this presentation, I focus on the polymeric nature of geopolymer.

A polymer is a macromolecule with definite size

and molecular weight. These two key values are
established by several complementary physical
methods working either in the solid state (electron
microscopy) or in solution (light-scattering). A gel,
on the other hand, designates an indefinite
amorphous compound with unresolved
dimensions.

Figure 6: Nano-particulate or geopolymeric micelle,

point of the arrow, after Kriven et al.

Kriven et al. (2003) used TEM (transmission

electron microscopy) to investigate the
microstructure of fully reacted potassium-
poly(sialate-siloxo) type geopolymers. It consists
of nanoparticulates ranging from 5 to 15 nm in
dimensions (50 to150 A) separated by nanoporosity, the features of which are of the order of 3 to 10 nm
(Fig.6). The nano-particulates represent a characteristic feature of the geopolymer matrix and their
dimensions suggest the presence of a macromolecule of definite size, and therefore, definite molecular
weight. It is the accumulation of these nanoparticulates, or individual particulates, that forms the
 -4-
geopolymer matrix. They are sometimes called precipitated particles and their dimensions are similar to
those of organic molecules.
The temperature stability of the geopolymer
nanoparticulates strongly supports the presence of
giant molecules. In other words, it is in favor of the
polymeric model.

Figure 7: Dimension of the geopolymeric micelle

(particulate) compared to colloidal silica, silica fume
and fly ash spheres.

Fig. 7 shows the very small dimension of this

geopolymer nano-particulate, when compared to
other spherical structures, colloidal silica, silica
fume and fly ash. The core of these
nanoparticulate geopolymers is made of
aluminosilicate frameworks that are similar to those
of rock-forming minerals.

Yet, there are major differences. In 1994, we

simulated a theoretical structure for K- poly
(sialate-siloxo) (K)-(Si-O-Al-O-Si-O) that was
consistent with the NMR spectra. It is displayed in
Fig. 8 and does not show the presence of water in
the structure. This is demonstrated by the fact that
27
Al MAS—NMRspectroscopy of all (Na, K)-poly
(sialate-siloxo) (Na, K)-(Si-O-Al-O-Si-O) showed
27
Al chemical shifts in the range of 55 ppm which
indicates that the aluminum is of the AlQ4(4Si)
type and is tetrahedrally coordinated, or more
exactly tetravalent (see Fig.17 below).

Figure 8: 3D-structural model for fully reacted K—

poly (sialate-siloxo), Davidovits (1994)

The absence of any other resonance in the 27Al

NMR spectrum and the extremely narrow peak at 55 ppm, excludes any residual singular building units of
low molecular weight such as dimers and trimers. (Na, K)-poly (sialate-siloxo) (Na, K)—(Si-O-Al-O-Si-O)
are true three-dimensional framework silico-aluminates with polymeric building units.

APPLICATIONS:

The applications are the result of a very precise chemical mechanism involving covalent bonding and
promoting the formation and production of ceramic-like materials, NOT JUST CEMENT and building
materials.


The atomic ratio Si: Al governs the chemical, physical and mechanical properties of ceramic-like materials
found in a wide variety of uses. Some geopolymer applications are still in development while others are
already industrialized and marketed. The following areas may be mentioned (Fig. 9):
- Si:Al=1: geopolymer ceramics and manufacturing techniques with low-CO2 greenhouse gas emission;
- Si:Al=2: geopolymer cements and concretes with low-CO2 greenhouse gas emission and low energy
demand; treatment and containment of toxic, radioactive and nuclear waste and mine tailings;
- Si:Al=3: fire- and temperature-resistant compounds for the manufacture of prototypes and tooling;
- Si:Al>3: high-tech geopolymer resins and binders for paints, coatings and grouts resistant to corrosion
and temperature; tooling for Aluminum industry;
- Si:Al>20: high-tech composites made of carbon fiber and others, resistant to fire and heat for
aeronautics and automotive, for the repair and reinforcement of civil engineering infrastructure. 

The wide variety of potential applications includes: fire-resistant materials, decorative stone
artifacts, thermal insulation, low-tech building materials, low-energy ceramic tiles, refractory items,
thermal shock refractories, biotechnology (materials for medical applications), foundry industry, cement
and concretes, composites for infrastructure repair and reinforcement, high-tech composites for aircraft
interior and automobiles, high-tech resin systems, radioactive and toxic waste containment, arts and
decoration, cultural heritage, archaeology and history of sciences.
 -5-

Figure 9: The atomic ratio Si: Al in the poly(sialate) structure determines the properties and application fields.
A low Si: Al ratio (1,2,3) initiates a 3D-Network that is very rigid. A high Si: Al ratio, higher than 15, confers
linear polymeric character on the geopolymeric material.

To sum up: GEOPOLYMER applications are NOT JUST FOR CEMENT. AAM are only
CONCRETE. Alkali-activated-material AAM is not GEOPOLYMER.

Part 3: (2016): AAM are not

polymers, so they cannot be
called “geopolymers”
I emphasize the fact that Alkali Activated
Materials (AAM) are not polymers;
therefore, they cannot be called
“geopolymers”. I present what scientists are
now writing about this issue. They now
agree with proven facts and that it is a big
scientific mistake to think AAM and
geopolymer as synonyms, and people shall
stop doing so. Geopolymer cement is not a
CSH derivative like NASH or KASH;
therefore, scientists are now stating that
applying the CSH terminology from Portland cement is not only inappropriate, but also calling them NASH
and KASH is totally wrong. Those who purposefully use and propagate these misleading languages
delude the understanding of the true chemical reactions that really occur (never a hydrate or a gel, but a
polymer), resulting in confused interpretations.

Got to: https://www.geopolymer.org/faq/alkali-activated-materials-geopolymers/ and on YouTube.

 -6-

Figure 10: Portland cement

chemistry vs Geopolymer
cement chemistry: a)
Hardening of Portland cement
through simple hydration of
Calcium Silicate into Calcium
Di-Silicate hydrate (CSH) and
lime Ca(OH)2; b) Hardening
(setting) of Geopolymer
cement through poly-
condensation of Potassium
Oligo-(sialate-siloxo) into
Potassium poly(sialate-siloxo)
cross linked network; c) 29Si
NMR spectra.

Fig. 10 compares the two systems: on the left, the hardening of Portland cement through simple hydration;
in the center the hardening (setting) of Geopolymer cement through polycondensation; on the right, the
29 29
corresponding Si NMR spectra. Si NMR spectra clearly show the difference in molecular structure. For
Portland, in the hydrated CSH, the silicon unit is of the type Si(Q0) for calcium silicate monomers, with
also some amount of calcium silicate oligomer of
types Si(Q1) and Si(Q2).

By contrast, for geopolymer, the structure is

polymerized into a silico-aluminate, three-
dimensional network of the type K—poly(sialate-
siloxo) wherein the silicon unit is of the type
Si(Q4).

Figure 11: Portland cement chemistry yields CSH

small hydrate molecule, and states Geopolymer
chemistry is based on NASH and KASH.

Portland cement chemistry starts with calcium

silicate hydration and end up with the so-called
calcium silicate hydrate CSH. Alkali-activated
cement materials scientists just substitute calcium
with sodium and potassium for geopolymer . They
claim that one gets sodium-alumino-silicate hydrate
NASH and potassium-alumino-silicate hydrate KASH (Fig. 11).

By doing so, they assert to understand

geopolymer chemistry, which is wrong. By arguing
that the end result is a simple molecule, a hydrate,
or a precipitate, they ignore the scientific facts
described in Fig. 10, namely the production of a
polymeric network .

Figure 12: 2016 International Conference on

Durability of Concrete Structures.

Other scientists are now recognizing this error.

For example, one team presented a paper at a
recent conference (Fig. 12).

The title reads (Fig. 13): Study on the Disposition

of Water in Fly Ash-Based Geopolymers Using ATR-IR. The title is a little misleading and seems to be
focusing on another issue. But the abstract written by the Australian-Chinese team provides clear
information which supports our present statement: “This paper addresses the question of whether the
main product of low calcium fly ash-based geopolymer is a hybrid namely sodium alumina silicate hydrate
NASH. The answer to this question is important for understanding geopolymer characteristics. ….”
 -7-

Figure 13: Paper title and Abstract of the paper

presented by J. Liu, Y. Fang and O. Kayali.

Their study demonstrates that Infrared

spectroscopy does not show the presence of
water (or hydrate). They write: “… The
assumption that the main product of geopolymer
is NASH is not favored …” This is a jargon that is
spoken in a politically correct language, yet, it
clearly means that the alkali-activated definition
with its NASH / KASH is false.

Fig. 14 summarizes our claim, namely that

geopolymers are high molecular structures or
macromolecules, in other words : polymers. On the
opposite, Alkali-activated materials AAM are based
on a low-molecular chemistry with small molecules
and are not polymers. It is therefore obvious that
they cannot be called Geo-Polymers. Geopolymers
and AAM are 2 very different chemical systems.

Figure 14: Why AAM are not Geopolymer?

It is a big scientific mistake to use both as

synonyms. Alkali-activation is a wrong terminology
for geopolymers.

In fact, the so-called alkali-activation is the first step

of geopolymerization. From a pure chemical reaction point of view we prefer using the most precise term
of “alkalination”. For the geopolymer preparation
process, nothing is left to chance. There are 6
precise steps displayed in Fig. 15.

Figure 15: The 6 steps of geopolymerization.

The preparation process follows the chemical

reaction and each step is connected with a process,
namely:
-1) Mixing (K-Na) silicate and metakaolin/or slag/or
fly ash.
- 2) Mixing time and mixing method.
- 3) Resting time.
- 4) Starting to harden.
- 5) and 6) Hardening.

Those who continue to mix up alkali-activation with

geopolymerization, wonder whether there are
other activators than their basic NaOH. The
answer is displayed in Fig. 16 and is self-explicit.

Figure 16: There is no activator, only hardener or

reagent.

There is no “activator” because we are dealing

with chemically reactive ingredients.

Geopolymerization does not require the entire

dissolution of inert or crystalline alumino-silicates.

To sum up: AAM are not polymers, so

they cannot be called “geopolymers”.
 -8-
Part 4 (2017): NASH / KASH is an invalid terminology
In 2016, a paper published by a group of scientists determined that there is no presence of NASH or
KASH in geopolymer cement (see part 3 above).

In this short excerpt, I explain this result by the true polymer nature of geopolymer chemistry. You will
learn what true NASH and KASH are, and in which context they are actually used. AAM and
geopolymer cement (wrongly shorten by some as “geopolymers”) are two very different and separate
chemistry (a hydrate/precipitate that is a monomer or a dimer versus a true polymer). None is a subset of
the other or its derivative which leads to confused interpretations.
Got to: https://www.geopolymer.org/faq/alkali-activated-materials-geopolymers/ and on YouTube.

Figure 17: Geopolymerization with intermediary oligo-sialate-hydrate formation, top part right, wrongly called
NASH/KASH by cement scientists. This hydrated molecule polycondenses into a fully reacted 3D geopolymer
network. Right, 27Al-MAS-NMR spectrum for K-poly(sialate-siloxo) K-PSS geopolymer.

Alkali-activated cement materials scientists just substitute calcium with sodium and potassium for
geopolymer. They claim that one gets sodium-alumino-silicate hydrate NASH and potassium-alumino-
silicate hydrate KASH. The hydrate terminology for calcium-based is correct, because the cation Ca++ is
practically insoluble in water. Therefore, the hydrate is chemically stable and does not generate leachates.
This is not the case for the cations Na+ and K+ which are very soluble in water. A hydrate such as NASH
or KASH will very easily leach out its cations Na+ and K+.

I have a beautiful example below which illustrates this behavior.

-9-

It is a US Patent issued in 2016

(Fig. 18), titled: “Alkali Metal Ion
Source with Moderate Rate of Ion
Release and Methods of
Forming”.

Figure 18: US Patent

US 9,340,465, B2, issued May 17,
2016.

It describes a new approach in

manufacturing potassium fertilizer,
starting with geological raw
materials (K-feldspar and the like).

The main claim of the patent (see in

Figure 19) is the production of KASH
gel, potassium-alumino-silicate-
hydrate, to make a fertilizer. The
cation K+ should be leached out by
water, when put in the earth and
help fertilize soils in a wide range of
countries.

Figure 19: MIT patent claiming the

manufacture of KASH/NASH for K-
fertilizer.

The terminology KASH is widely

used here by the applicant who is
the MIT, Massachusetts Institute of
Technology, Cambridge, Boston.
This is logical here. MIT is not claiming that
KASH is AAM, nor geopolymer.

But those material scientists who make

AAM and NASH are producing a material
which contains soluble alkali Na+ that will
leach out, like in Figure 20.

Figure 20: AAM block with NASH

efflorescence (white blooming).

On the opposite, those material scientists

and civil engineers who follow
geopolymerization precise process and
chemistry are manufacturing advanced
geopolymer concrete, like in Figure 21.

Figure 21: Geopolymer concrete (2015).

 -10 -

Afterword:

24 years ago, I made a presentation at the First International Conference on Alkaline Cements and
Concretes, titled: Properties of Geopolymer Cements, in Kiev, Ukraine (1994). It was there, at Kiev, that V.
D. Glukhovski developed in the 1960-1970s the concept of alkali-activated materials which he had coined
“Soil-cements”. The excerpt below was already discussing the issue, "AAM are not Geopolymers".

The full text is available here in Research Gate and also at the Geopolymer Institute Library, at
https://www.geopolymer.org/category/library/technical-papers/ , paper #8: Alkaline Cements and
Concretes, Properties of Geopolymer cements.

(……)

TERMINOLOGY: GEOPOLYMER VS. ALKALI-ACTIVATED ALUMINO-SILICATE

We mentioned above that, because alkalis are generally thought of as the cause of alkali-aggregate-
reaction, the present tendency is to avoid any addition of alkali in ordinary Portland cement. According to
the terminology generally in use by cement scientists and concrete experts, both cementitious systems
discussed above in terms of the relationship between alkali-aggregate-reaction and 27Al MASNMR
spectroscopy, should be named «alkali-activated cements» [47]. Yet, the cement which resonates at 55
ppm, AlQ4(4Si) building units (Fig.17 and Table 8), does not generate any deleterious alkali-aggregate-
reaction (Fig. 15, ASTM C227 bar expansion test), whereas the cement which resonates at 66 ppm,
AlQ3(3Si) building units, reacts with susceptible silica. In addition, in the first cement category, the SiO4
tetrahedrons are of the three-dimensional cross-linked framework (Q4) type, SiQ4(3Al)-SiQ4(2Al)-SiQ4(1Al)
(Fig. 16), and are essentially different from the Si(Q0), Si(Q1), monosilicate and disilicate structures of
calcium silicate hydrates, which are the main constituents of the second alkali-activated Portland cement
category. It becomes obvious that the terminology in use generates confusion and is a severe obstacle for
any further beneficial scientific and commercial developments of alkali cementitious systems.

The concept of Geopolymer and Geopolymerization is well accepted in the science and technologies
involving advanced materials. Geopolymers result from the polycondensation of polymeric alumino-
silicates and alkali-silicates, yielding three-dimensional polymeric frameworks. Cement scientists should
admit that cements involving alkali contents of 9.2% and higher, which do not generate any dangerous
alkali-aggregate reaction, cannot comply with the existing codes and guidelines and should get a
distinctive appellation. To call them Geopolymer cements or Geopolymeric cementitious compounds or
Poly(sialate) cements, focuses on their unique properties without being confused with regular alkali-
activated Portland cements. (……) End of excerpt.