At the stage of unstable phases, owing to weak bonds of water molecules that are depolarized with chlorine ion and
weak bonds of the reaction products, nitrate ions come to react, and the sequence of these reactions is determined by
their inherent chemical activity, alkali level of the solution and the intermediate reaction product - calcium aluminate -
with which the following dissociation reaction is most likely:

  3Ca(OH)2 + 6NaNO3 + 32H20 + Ca3(AlO3)2 3Ca3(AlO3)23Ca(NO3)2·32H2O + 6NaOH      (4)
                                                                                                                         ↓    ↓             
                                                                                                                       Na+  OH-
This reaction yields a low-soluble double salt of calcium hydronitroaluminate with an increase in pH of the pore fluid.  The stability of reaction (4) is insured by an almost simultaneous reaction of sodium sulfate.  The consumption of starting components for another reaction (5) result in their shortage and in a one-way character of dissociation:

3Ca(OH)2 + 3Na2SO4 + 31H20 + Ca3(Fe2O3)23CaO Fe2O3)2CaSO4·31H2O + 6NaOH(5)  yielding calcium hydrosulfoferrite.

Therefore, if such an electrolyte is added at a concentration that insures a change in solubility of mineral binders without reacting with them, with a subsequent formation of hardly soluble complex compounds - calcium hydrosulfoaluminate, calcium hydrosulfoferrite, calcium chloroaluminate and tricalcium chloroaluminate - from the resulting solution, the overall volume of the crystalline component of the structure increases all at once in parallel with normal concrete cure. Moreover, a protracted reaction allows the ion force of free water (which later is the pore fluid) so as to form saturated solutions and to form additional double hydrate salts.
Calcium-containing electrolytes accelerate hydration and hardening of silicate phases of cement owing to a higher probability of formation of three-dimensional germs of a new phase.  These electrolytes also disperse the products of hydration through dissociation with anion-kation groups:
                                           +         +    + --                          
         Ca(OH)2 + Na2CO3 CaCO3 2Na 2OH                                                            (6)

The above-described processes insure a high hardening rate and a fast rise of strength of the protective layer by a better use of the potential of allite 3CaO SiO2 C3 S.  Chloride ions that are still in the liquid phase are products of displacement.  They form solvation shells at the boundaries of cation fields thereby preventing free calcium from leaving the structure-forming reactions.  At the same time, nitrate and ferrite ions accumulate in the free water polarized with chlorine ions to form solutions of increasing ion strength.  These solutions will, in turn, accelerate hydration of allite.  The manifest relay-like character of these processes allows allite to develop to a greater extent into a symmetrical three-dimensional conglomeration with isotropic properties.
The above-described processes insure the adhesion of the protective layer to the surface of concrete, the protection of reinforcement against corrosion, density and low permeability of structure to a depth of penetration of electrolyte into the body of concrete being protected.

Dr. Alex  Rusinoff, Ph.D,S.P.

KALMATRON® KF-B protective composition is designed in 1982 to 1994 for the restoration of strength, impermeability and corrosion resistance properties of any cementitious structure. The product is patented in the USA under the #5,728,428

Maintenance and restoration of concrete structures is a new field in the concrete science.  It is almost a medical type of problem of structure regeneration that is emerging at the interface of physics and chemistry. It is the stage at which first hypotheses, practical solutions and numerous discrepancies in the understanding of the same uncontrollable processes of concrete maturing and aging would come to life. And no sooner the scientists had managed to find the next “philosopher’s stone” than it became clear that it could help in redefining the problems and were only good for a given application [1; 5].

This paper is not an exception. KALMATRON® KF-B developed within the range of capabilities inherent in its formulation. The gist of this formulation is a simple relay sequence of reactions of cement stone phase shift: penetrating the old concrete, dissolving all that can be dissolved to any possible extent, spreading out within concrete voids, and hardening as one solid. And this is the real distinction from other products, even they have claim the same brand and properties.

It is known [3] that concrete does not have enough time to achieve complete maturing of its structure. This time is available during the relay sequence reactions, beginning with cement hydration. Restoration of concrete is based on dissolution of starting meta-stable binders. The rate and extent of dissolution depend on energy performance of a solvent and degree of breakage of bonds between the binder and aggregate. 

Dissolution products form supersaturated solutions from which more thermodynamically stable new hydrate formations are built. These new hydrate formations are characterized by a lower solubility, greater surface area and higher density [4]. It is clear that it is not possible to provide a universal protective composition because concrete would have an unpredictable degree of wear depending on its initial properties, operating conditions and age by the moment the repair has to be performed. Since it does not seem possible to estimate with a statistic confidence even the ratio between crystalline and amorphous phases in cement stone of real structure, one cannot provide a calculation theory to estimate the overhaul interval with a fair degree of probability for non-cyclic processes [1;4].
Requirements Imposed Upon Protective Cover Compositions

1.  Compatibility with the material of a structure to be protected.  The composition should become property of the 
    material and cannot      depend on its initial characteristics.
2.  Functional permeability to assure moisture and temperature equilibrium between concrete or mortar of the  
    structure and the ambient environment, while providing a structural barrier against corrosive media.
3.  Uniform distribution of properties of a protective composition over the area or body of a structure to be protected.
Impermeability and Resistance of KALMATRON KF-B to any Aggressive Media

The concept of impermeability of capillary and porous bodies is normally associated with the maximum possible pore filling both on the surface side (daubing and adhesive proofing) and within the pore system (guniting and filling composition).  However, with the complete insulation of the capillary and porous system, the values of osmotic and crystallization pressures (7) that are as high as 2·107 N/m2 (200 Atm) in concrete containing structural moisture may cause catastrophic consequences if concrete voids are not open into atmosphere.
The same applies to the filling methods if the compositions used for filling harden at once as is the case of gypsum compositions or silica.  For this reason, the time and depth of concrete structure penetration with KALMATRON® KF-B are so important.
The effect of osmotic pressure during diffusion of dissociating salts in cement stone described in the general form by Vant Hoff equation for osmosis (11).

  P  =  n / w · R T  ln | Va  / Vb | ln | Ca / Cb |  ln | T |  = μ  ln | Va / Vb |  ln  | Ca / Cb ln  | t |;   (7)
n- is the number of moles of insoluble salts;
w- is the solvent volume;                        
μ - is the chemical potential
R - is the universal gas constant;              
Va / Vb - is the rate of initial to final water content
T- is the ambient temperature;   
Ca / Cb - is the ratio of initial to final solution concentration.                     

Elementary unit of concrete volume with self tensions created by osmotic 3 D- forces:  
Px = m ln|Vh/Vf| ln|C1/C2|ln|t| = 0, wherein C1=C2;                                       (8)
P y  =  m  l n   | Vh / Vf   |  ln   | C1 /  C2  |  ln   | t  |  <  0;   since C1 < C2;      (9)
P z   =  m  ln  |Vh  / Vf  | ln  | C1 /  C2  |  ln   | t  |  >  0,  because  C1> C2.     (10)

Therefore, the possibility of separation of the solution according to areas of molecular penetration (for water and weak solutions) and for ion penetration assures a deep penetration of concrete structure with KALMATRON® KF-B. For concrete with a density below of 2200 Kg/m3 the penetration depth is at least 150 mm, and it ranges from 10mm to  50mm for concrete with density of up to 2400 Kg/m3. This effect allows the conventional preparation of structures, involving cleaning up to a structurally perfect depth, to be dispensed with. It only takes to wet the surface with water.  

KALMATRON® KF-B solution diffuses with the inter-structural water, which is a diamagnetic fluid, will loose the dipolar orientation of molecules as a consequence of dissociation of charges of opposite polarity.  In other words, the diamagnetic elasticity of water disappears.  This will result in a weakening of the water thrust effect upon the pore and capillary walls (the electrolytic relaxation of the cement stone structure), and concentration of the solutions will rapidly come to an equilibrium. Strength “decreases” during tests, and permeability of concrete remains unchanged since pores are not yet filled with new formations, and the reaction pressure in the pores is equal to zero:

Pxyz = m ln|Vh/Vf| ln|C1/C2|ln|t| = 0, wherein C1=C2                            (11)

“Negative” strength values in 28-day tests are, therefore, no more than a part of the process during which the actual concrete strength is gained.  Further observations show a 25% to 30% strength increase. During diffusion the salts are accumulated in gel pores, concentration of structure solutions C2 increases to cause an osmotic
compression of the structure with a negative pressure Pk.  In this case, permeability and “strength” are improved since the reactive pressure on the pore walls is negative:

Pxyz = m  l n   | Vh / Vf   |  ln   | C1 /  C2  |  ln   | t  | < 0; since C1 < C2   (12)                                                                                           
Concentration is leveled out to the state (12) in area 3 where an active recrystallizing takes place, but dissolution of unstable phases including poorly soluble salts concurrently occurs as follows:

         Am Bn mA + nB                                                                            (13)

From this moment, when kation A has been replaced by kation B, the dissolution of the initial binder is completed, to cause an increase in concentration of the initial solution C1, and the osmotic pressure will then develop tensile stress in cement stone structure to result in a material “strength increase” of up to 50% and a decrease in permeability, because the density of pore filling also becomes greater:

Pj = m  ln  |Vh  / Vf  | ln  | C1 /  C2 |  ln   | t  | > 0,  because  C1> C2   (14)

The most dangerous are apparently the boundaries between the areas that are subjected to the effect of tensile forces from mutually opposed reaction forces.  It is for this reason that the use of unbalanced compositions to improve concrete permeability results in destruction in 3 to 5 years.  Similar to high-density insulating layers, they are also dangerous from the point of view of reliability because of the uncontrollable effect of ambient temperature. The exponentially expressed temperature ln | t | in (14) will sway the structure season by season to a sudden collapse.
Stability of concrete characteristics is due to an unstable equilibrium of mutually strained phases of cement stone.  For this reason, short-term data on an increase in strength of immature structures obtained in testing concrete with any additives cannot be relied upon because the force the osmotic and crystallization pressures       have opposite effect during different periods that depends on hygrothermal equilibrium with the ambient environment. A weak correlation between strength and permeability of concrete, as shown above, facilitates operations aimed at reducing permeability. 
      The problem is to estimate the difference between osmotic pressure values at the phase boundaries and rates of formation of crystal areas in concrete(Fig. 5). This is accomplished by oversizing the volume of the gel component of cement stone with ions of strong electrolytes that decoagulate the gel so as to create the sieve effect for ion and molecular solutions.                                                                      
      Impermeability to liquids with high parting surface such as alkalis, acids and oil products can be assured either by presence of gases in the capillary and porous system of concrete (Jamen’s effect) or by making the liquid surface area comparable with the total surface area of the whole system. Therefore, the type of a new formation crystal in the concrete pores is more important, and the degree of density of pore filling is not so critical, because the wetting surface area has a stronger effect that the extent to which pores are filled.

      Formation of calcium hydroxochloroaluminate (acicular crystal hydrates) and sodium hedroxocholoro-aluminates (laminar crystal hydrates). The development of these crystals along the optical axis within concrete pores is not possible because their strength is low at the maturing stage, and they are either broken down at the  contact with the pore walls or change the direction of growth which is important for durability and safety of the restored structure.

It has been shown in the crystallization theory that the rate of formation of crystal nuclei within a unit of volume depends on the  degree of oversaturation C1/C2.  At the same time, it also depends to a no lesser extent on the specific interphase energy m that is defined by the chemical potential (16) of the entire thermodynamic system:

μ = n / w·RT ln | Vh / Vf | < 0;   Vh << Vf,                    (15)

the value of which is fairly controlled by the quantity of moles of dissolved electrolyte salts n and the volume of pore moisture Vh. By putting (15) to (14), the value of stresses built up by the osmotic pressure and crystal nuclei obtained from dissolved products of the clinker in the old cement stone can be easily estimated.  The comparison of calculations with the experimental data showed a narrow area of admissible ion force values for concrete and allowed a correlation to be established between the chemical potential of an aqueous solution of the protective composition and permeability of concrete. 

KALMATRON® KF-B proved as the most reliable in making concrete less pervious when the concrete has either a poor development of the pore structure or defects of load, corrosion or temperature and shrinkage origin.  It should be noted that, since this composition has a high concrete penetration capability, the integrity of the cover layer is not an imperative, which is important especially for structures in which the surface is exposed to destructive mechanical or hydraulic factors.
After integration of (16) for two variables μ1 to μ2 and s1 to s2:
Dt = -0.5•(m²2 - m²1 )•ln|s2 /s1 | • ln|C2 / C1| • ln|T| = -0.5 • (m²2 - m²1 )• E• ln|d| • ln|C2 / C1| • ln|T|;  (17)
wherein - d - are the absolute deformations of structure shrinkage;
        E - is modules of elasticity of concrete.

Any theoretical studies of the mechanism for estimating changes in density D involve shrinkage d. 
The shrinkage is caused by a loss of water from concrete mix through hydration and evaporation:

            |m/ μRT + Ln |Vf| 
Vh =  e                                > 0                                                                                          (18)


Vf- is the total amount of water in concrete mix;
Vh- is the water residue in concrete structure;
m- is the mass of chemical additive.

1. S.V.ALEXANDROVSKY,  The creep of concrete. Moscow. 1975

2. P.H. EMMONS, A.M. VAYSBURD, Long-term durability of concrete repairs under severe environments, Structural  
  Preservation Systems,  Inc., Baltimore, MD, USA, pp. 709-716,.

3. P.C. KREIJEGER, Inhomogenity in concrete and its effect on degradation: a review of technology, Protection of 
   concrete, Conference of Dundee,Scotland,UK,p.p.31-52 (11-13 September,1990) 

4. A.M. NEVILLE, Properties of concrete, New York, 1993.

5. A.M. VAYSBURD, Some Durability Considerations for Evaluation and Repairing Concrete Structures, Structural 
  Preservation Systems, Inc., Baltimore,pp.29-35,(1993).

6. G. VERBECK, Cement hydration reactions at early ages, J.Portl.Cem. Assoc.Research and Development 

7. A.RUSINOFF, The phenomena of osmotic oscillator by absorption  of the atmospheric salt solutions by surface 
   layers of exterior   building walls. Collection of scientific works. Khabarovsk Railway Engineering Institute, Russia, pp. 59-63, (1988).

8. A.RUSINOFF, Exterior Walls In Extreme Climates, Protection  of concrete, Conference of Dundee, Scotland, UK,  p.p. 541-547  (11-13 September, 1990).


When KALMATRON® KF-B is mixed with water and applied to a surface of concrete being protected, a number of consecutive and simultaneous reactions take place between the components of the composition and between them and cement components as follows:
1.CaO + H2O Ca(OH)2
2.Ca(OH)2 + CaCl2 →↓ Ca(OH)Cl2 + CaOH
3.Ca(OH)2 + NaNO3 →↓ Ca(OH)NO3 + NaOH(1)
4.Ca(OH)2 + Fe2SO4 →↓CaSO4 + 2FeOH
5.Ca(OH)2 + K2CO3 CaCO3 + 2KOH

Free calcium oxide of cement forms calcium hydroxide when mixed with water (1.1).  Calcium hydroxide then takes part in exchange reactions with sodium nitrate and calcium carbonate and sulfate and with calcium chloride of the protective composition to form low-soluble and hardly-soluble acicular crystals of hydroxonitrates Ca(OH)NO3 (1.3) that will continue to grow well after completion of structure forming of cement stone by using free pore water and Ca ions released from cement stone gel.  These crystals have a micro-reinforcing effect on segregation within voids under the effect of temperature, shrinkage and corrosion.  Therefore, a primary structure reinforcement framework is formed within the protective layer as early as at the setting stage.  This framework is built up in the direction of mass transfer of a diffusion flow, i. e., in the direction deep into the pore system of the body of concrete being protected.
Hardly soluble double salts of calcium sulfoaluminate 3CaAl2O3CaSO4·31H20 are crystallized at the same stage.  The crystals are in the form of hexagonal syngonite-like structures or a package of parallel laminae with interstices filled with intercrystalline solutions.  The density, volume and strength of the entire package depends on density of such solutions.  When moisture gets into the interstices from the ambient medium, the solutions are diluted, and the package volume increases.  Given the conditions in the pore space of concrete, this is the explanation of an exponential decrease in permeability with time during tests.  If temperature decreases, the intercrystalline solutions break into crystalline hydrates and solutions of residual concentration.  The volume of the interstices decreases, and density and strength of structure as a whole increase to ensure a high frost resistance.
During a further maturing stage, low-soluble double salts of calcium nitrochloroaluminate 2CaOAl2O3Ca(OH)Cl2·10H2O and nitrochloroferrite 2CaOFe2O3Ca(OH)Cl210H2O are formed on the primary framework in the form of the same hexagonal syngonite-like structures.  However, concentration of intercrystalline solutions is so high that their density does not almost change with an inflow of moisture from outside.  High level of molecular bonds is explained by the effect of chlorine ions upon dipolar water molecules.  This phenomena is similar to the case where water is magnetically treated before mixing concrete to improve concrete strength.
Adding chlorine ions to the compounds dissolved in water has a polarizing effect on dipolar water molecules to lower the level of molecular bonds of water.  Owing to weak bonds in the presence of calcium hydroxide, an alkali group is released into the water to protect calcium against dissolution at the maturing stage:

  3Ca(OH)2 + 6NaCl2 + 30H20 + [3CaOAl2O3]3[CaOAl2O3CaCl2·10H2O] + 6NaOH    (3)
                                                                                     ↓                   ↓      ↓
                                                                                            Cl-          Na+  OH-

Tricalciumalumochloride formed as a result of reaction (3) forms hardly soluble solid phases when water is released for simultaneous hydration reactions.  The alkali and the internal pore moisture form solutions inhibiting metal corrosion that also have a low eutectic temperature of -126°F (-70°C) at the stage of a stable phase condition of cement stone.
Removed 3 mm of KF-B layer
20 mm of KF-B penetration
Figure 1. The cylindrical core concrete specimen with removed 3 mm of KF-B layer on a top. Seen 20 mm of darker zone of regenerated concrete as a result of KF-B penetration.
Figure 2. The fragment of regenerated concrete structure as a result
               of KF-B penetration wherein:
A - Dissolution of unhydrated cement grains at the 7-th day
     after KF-B applied.
B - Sewing effect is detected by melted-like crak adges.  

  When the solutions precipitate to the pore walls, acicular crystals of calcium hydrochloroaluminate  and laminar  sodium hydroaluminate crystals are formed. The development of these crystals along the optical axis results in micro-cavity edges growing together.  This is a so-called “self-healing” of defects in the concrete structure (Figure 2 )  
Figure 3. Studying cracks into concrete specimen after thermical chock. The compressive strength is fallacy high, but imperme-ability degrading very fast.  X60  
Figure 4. Early stage of KF-B healing characterized by rounded, or like melted looking shapes. The compressive strength is not been changed on this stage, but impermeability is higher by 2 times. X60   
Figure 5. Later stage of KF-B healing with melted looking crystalline new-growths "A" and opened shells of regenerated cement grains "B". Compressive strength, density, and impermeability are highly performed after 14 to 28 days. X600
Figure 6. Later stage of boundary crack healing by KF-B. The resistance to rupture, density, and impermeability are highly performed after 3 to 7 days.
X60 - X600
Figure 7. Cylindrical mold-holder with mechanical indicator of deformations for evaluation of osmotic-crystal pressure into the concrete structure. The maximum osmotic-crystal pressure was 6 MPa, which might be compared with rupture strength of tested concrete is 3.5 MPa.
Figure 9. Evaluation of concrete crepe as an indication of stability of concrete characteristics under salt condition. Demonstrates the fallacy high compres-sive strength during of concrete deterio-ration, which explains the reason of sudden collapse of massive structures.
Figure 8. Rusinoff's Osmotic Oscillator station is used for evaluation of osmotic pressures during of salt solution penetration through the cement plate.
Studied the principal of concrete collapse under cycling deformations of osmotic diffusion into cementitious matrix.
Figure 10. Cylindrical concrete specimens with mechanical indicators of deformations for evaluation of osmotic-crystal pressure into the concrete structure and distribution of inter-porous solutions. The results of this 2 years experiments approved effectiveness of KF-B application and gave correct vision on a concrete durability problem.  

The present review of capabilities of KALMATRON® KF-B protective composition reflects physical concepts of repair and restoration of concrete that require a certain generalization.  The experience of structure protection has been unique so far because of the lack of a systemic approach to the development of this field [2].  The same problem exists when making choice among protective compositions for which the field of application is determined based upon the results of standard laboratory tests or by making a comparative analysis of properties of several compositions.

At the same time, this is also inadequate since an acceptable choice has to rely upon actual condition of a structure that can be assessed by examining for wear and reparability.  However, the most important point resides in the fact that, in spite of the existent prejudices in the construction industry based on belief in perpetual properties of concrete, the law of energy conservation is also true in this case:

1. An increase in durability of restored concrete structures can only be assessed at the time of a specified overhaul period.  Similar to any heterogeneous body, concrete
repels foreign inclusions.  This is especially true of polymer reinforced concrete, fiber reinforced concrete and concrete with any insulating coating and non-zeolite additives.

2. An increase in compression, flexural and tensile strength caused by the use of any composites in concrete mix occurs only during cement stone equilibrium phases.  This equilibrium can only be disrupted by a change in ambient temperature, humidity or dynamic load.

3. Lower permeability comparable with durability of concrete can be achieved if a maximum possible wetting surface is available within the structure.  An excessive pore filling can result in high pressure within the structure and early decay of the structure.

4. Various generalizing concepts and assumptions of the theory of concrete strength are based on a fair and long-term statistics of tests and observations.  However, in the development of theoretical support for durability and restoration of properties of concrete, an almost biological model of its behavior is emerging, especially taking into account the environment involving half-permeable membrane shells [1], independent electroosmotic processes [7], diffusive catalysis [3], regeneration and association of elements, and elementary and microscopic new formations that are analogous to processes occurring in an organic cell. 

The difference is in a very low rate of change.  This is the only reason why the mechanical application of concepts of the theory of strength to the problem of concrete durability leads to a dead end, with random results and fragmentary vision of the processes.

The provision of KALMATRON® KF-B composition and related processes represents a practical trend toward restoration of old concrete that has been practically implemented to the largest extent.  It is obvious that the capabilities of this composition are limited by reparability of concrete, and the composition cannot bring solution to all problems of maintenance and restoration of concrete.

At the same time, practical results support the advantages of this development beyond any doubts with an outlook to its further adjustments to the industry requests.
Figure 11. Concrete toroid specimens in the 2-point holders for monitoring of phases transformation, self-healing, sorption and exothermic distribution, and speed of aging under different aggressive liquids. 
Figure 12. Closed voids created by gases during the dough phase.
Figure 13. Fresh KF-B paste bordered with concrete structure. X60
KF-B layer
Figure 14. Diffusional boundary between KF-B paste and concrete structure at the age of 7 days magnified by X600.
Figure 15. The same diffusional boundary between KF-B paste and concrete structure at the age of 7 days magnified by X60.
Figure 16. Complete recrystallization of diffusional boundary between KF-B paste and concrete structure at the age of 14 days magnified by X600.
Figure 17. Crystallized micro voids of concrete structure by KF-B new growths at the age of 14 days. X60.
Figure 18. The same crystallized micro voids of concrete structure by KF-B new growths at the age of 28 days. X60.
Uniformity of KALMATRON® properties to Concrete
Figure 11.a. Concrete toroid specimens in the 2-point holders for monitoring of phases transformation, self-healing, sorption and exothermic distribution, and speed of aging under different aggressive liquids and different levels of concrete sorption ability. 
Isotropy of concrete is not possible because dehydration is accompanied by decomposition of structure from the interior toward the outer surface over the phase spectrum from solid to gases during the whole structure operation time.  Therefore, the destruction occurs in a kind of a layer by layer fashion, moving inwardly from the surface. 

This phenomenon, which is rather unusual for concrete, allows scientists to develop various facilitated restoration methods.  For the rest, the dependence of concrete on ambient conditions and on the level and nature of load does not leave any room for doubt about usefulness of the generally adopted concepts and assumptions.

Any rheologic characteristic can be used as the parametric measure of uniformity, but the most convenient property for the practical use is density of concrete.  Equation (16) gives a general idea of the parametric dependence of density on processes of natural maturing or forced post-maturing of structure:

         Dt = -  μ/ s • Ln|C1/C2| • Ln|T|;                                   (16)

Dt - is the density of structure vs. time (kg/cm3);

μ- is the chemical potential of an aqueous solution of chemical additive (J);

s - is the stresses created by surface tension forces within concrete mix or by crystalline new formations within concrete  
structure (kg/cm2);                                                           

C1- is quantity of a binder in concrete mix (kg/kg);  C2- is the concentration of chemical additive in concrete (kg/kg);

T- is the ambient temperature (Kelvin).
With a decrease in water residue Vh, films of hydration products begin to appear around binder grains. These film-forming compounds in cement are calcium hydrosulfoaluminate and calcium hydrosilicate.

The formation of shells of very fine hydrated products is the primary reason for the leveling out of hydration rate.  According to (18), the volume of hydrated product is greater than the volume of an equivalent amount of the starting product, and stresses are built up within the resulting shell around the grains so as to cause a partial destruction of the grains.  This results in shrinkage of cement stone.  Water gets into the interior or damaged shells to dissolve non-hydrated binder and to form new crystalline bonds between the particles and new crystalline films around non-hydrated binder.  This process can be traced owing to a triggering mechanism of stress relaxation.

Therefore, the shrinkage as variation of cement stone dimensions is the result of a consecutive effect of two processes: crystallization of new formations that leads to a periodic change in volume and a process of destruction of crystalline contacts and films that results in a periodic decrease in volume or shrinkage.
It is not possible to level out the rate of hydration, i. e., to retain monotony of equation (18) since the initial ion force of a liquid solution will always be stronger than the ion force of phase products which, according to (18), is the cause of appearance of interstructural “negative) density such as cracks, cavities and separation:

                                    m1   > m2    -- D;                                                                                  (19)
It follows from (17) that a linear increase in density can be achieved by a simultaneous increase in the ion force of the solution m and a decrease in surface tension forces s.  However, this is a philosophical problem of the cause and consequence, rather than a practical problem of the physics of solid. It follows from (17) and (18):
                              ln | d | = - 2• D / (m²2  - m²1 )• E • ln | C2  /  C1 | • ln | T |,                                (20)
which dictates the aim of achieving weak shrinkage deformations through an increase in the crystallizing component over the phase spectrum of the concrete structure.  This is accomplished through an increase in structural density D of concrete owing to the completeness of the hydration reaction.  This equation is true in practice for high aluminate cements with a perfect fineness of grinding of maximum 0.5 to 0.09 Å. 

For medium and low aluminate cements and for coarsely ground cements, various laboratory-obtained correction factors have to be used which is often impractical.

To summarize, the problem resides in the rate and completeness of cement grain hydration.  Equations (1) through (4) describe dissociation reactions weakening the bonds in primary colloids of potassium and sodium hydro- chloro-aluminate.  This allows water influx through hydrate films to non-reacted grains of cement to be prolonged so as to result in the hydration volume being increased by 20% to 30%.  Shrinkage is reduced to a similar degree in comparison with conventional concrete.
Inventions of the 21-st Century