Repeatable effects
To test the reliability of the device, in standard operation, our test has been repeated in summer and winter for the pure water solution and polyacrylamide solution, respectively. The f is 15hz-50hz, and the results are shown in Fig. 2, and the curves tested for the same kind of samples are overlapped, which indicates that the device is reliable. For the polymer solution, the molecular weight Mv=2.97×106g/mol, θ = 300ppm PAM solution has a flow distribution in the range of 3-12m3/h, which lies below the curve of the pure water solution, showing a drag reduction effect.
Concentration effect
The f is controlled at 15-50hz and the T = 24–28°C in loop pipe. Solutions of polyacrylamide with θ of 25, 50, 100, 150, 200, and 300 ppm are tested and the results are shown in Fig. 3. The polymer molecular interaction diagram is shown in Fig. 5 (a) and the conclusions are as follows: (1) The DR curves of polymers with different Re first increase with increasing θ in the early stage. Before the saturation concentration, the polymer molecules are relatively dispersed in the solution, thus an increase θ will increase the number of effective molecular chains for drag reduction; (2) After reaching the saturation concentration, the DR starts to decrease, according to the concentration effect. Meanwhile, the higher the Re is, the larger the corresponding saturation concentration is. This is because when a polymer concentration is greater than the saturation concentration, the long polymer chains become entangled with each other, requiring a stronger turbulent force to rearrange the polymer molecules to perform drag reduction. Therefore, there is an optimal operating polymer condition, in which the polymer can play its better drag reduction effect.
Reynolds number effect
Our tests are carried out on solutions with θ of 25, 50, 100, 150, 200, and 300 ppm at T = 24–26℃, Re ranged from 15000 ~ 120000 in a turbulent flow. The test results are shown in Fig. 4 and the interaction of the polymer chains is represented in Fig. 5 (b): (1) The maximum drag reduction Reynolds number(Remax), has been reached for polymers at θ = 25-50ppm in the initial test case, where the polymer long-chain molecules are fully stretched, with a maximum DR of 37%. As the Re increases, the fluid turbulence and the recirculating flow through the pump cause some of the polymer long-chain molecules to be fractured, which results in the DR decreasing. The lower θ is, the more polymer molecules are fractured and the greater the slope of the curve is; (2) The long chains of polymer molecules at θ = 100-300ppm are initially entangled with each other. The higher θ is, the more effective the long chains of polymer molecules are after being stretched, and the higher the slope of the curve is. After being stretched, the polymer molecules will be shared by the polymer molecules through the pump. The viscosity of the solution will decrease, and the drag reduction effect of the long-chain polymer molecules will be lost. The Remax=78000–80000, and the higher the concentration is, the larger the Remax is. When Re > Remax, DR starts to decrease.
Molecular Weight Effect
In the loop pipe, PAM solutions with different molecular weights at θ of 100 and 300 ppm have been tested, within f = 15-50hz and T = 24–28°C. The results are plotted in Fig. 6 and Fig. 7. From Fig. 6, it can be concluded as shown below that at θ = 100ppm, the DR of the polymer increases as the molecular weight of the polymer increases, showing a positive correlation [41]. As shown in Fig. 6 (a), taking the critical flow rate Qc=7-8m3/h, when the flow rate Q < Qc, the long-chain polymer molecules can give full play to their roles, and the DR increases; when the flow rate Q > Qc, the molecular weight of the long-chain polymers is broken by turbulent flow and replaced by polymers with low molecular weights that have limited drag-reducing ability, and thus DR gradually decreases. Converted to Re, as shown in Fig. 6(b), polymer solutions with high molecular weights have higher apparent viscosities, and thus higher DR can be achieved at lower Re.
Comparing with Fig. 6(a) and Fig. 7, the results of PAM solutions with θ of 100 and 300ppm, the conclusions are as follows: (1) Increasing the θ of low molecular weight PAM solution can significantly improve the DR. The initial drag reduction rate of θ = 300 ppm Mv=6.63×105g/mol PAM solution is 8.7% higher than that of θ = 100ppm PAM solution, and the drag reduction rate can reach 27.7% at the flow rate Q = 7m3/h. This is because the increased concentration allows more polymer long-chain molecules to play the role of drag reduction in turbulent flow; (2) Increasing the θ of PAM solutions with high molecular weight causes the long-chain polymers to become entangled with each other, increasing the apparent viscosity and requiring some resistance to separation. Therefore, the higher the molecular weight at the initial test stage is, the greater the reduction in DR, which is 3.7% and 1.3% for Mv=1.50×106g/mol and Mv=9.86×105g/mol PAM solutions at Q = 3.5 m3/h, respectively. Then, along with the increase of turbulence intensity, the long-chain polymer molecules are untangled, and the DR gradually increases, the higher the molecular weight is, the higher the slope of the curve is, and at the flow rate of Q = 12m3/h, the DR of PAM solutions with molecular weights Mv=9.86×105 g/mol and Mv=1.50×106 g/mol is increased by 9.7% and 11.5%, respectively.
Shear resistance behavior
To evaluate the shear resistance of PAM solutions with different molecular weights and different concentrations, in our experiment, we take PAM samples at f of 15hz, 20hz, 30hz, 40hz, and 50hz for the viscosity test, and the results are shown in Fig. 8. Under the same shear situation, the viscosity of PAM solution decreases gradually with the increase of shear rate. The larger the molecular weight of the solution is, the greater the slope of the curve is.
After the completion of the first test of the PAM solution, the test has been repeated at an interval of half an hour and the results are shown in Fig. 9. Our conclusions are as follows: (1) In the first test, Remax is observed, which the DR declines as the Re exceeds Remax. Conversely, in the second test, an increase in DR correlates with a continuous enhancement in Re, with the curve ultimately leveling off; (2) When θ = 100ppm, the higher the DR of the polymer solution with high molecular weight, the stronger the shear resistance of the polymer, and the smaller the slope when the curve is reduced; (3) Compared with the first test, the DR is significantly lower in the second test, and the gap between the two test curves decreases as the Re increases, and the larger the molecular weight, the smaller this gap becomes. The difference in DR for PAM solutions with molecular weights Mv=1.50×106g/mol and Mv=9.86×105g/mol is 13.7% and 27.8% at Re = 30,000 and 4.4% and 4.7% at Re = 100,000, respectively. This is due to the first test when subjected to mechanical shear action, the molecular chain is easy to break at the point of rapid fracture. The number of molecular chains increases, and each molecular chain subjected to shear force becomes smaller, the second test then continues to break the chain. For these reasons, high molecular weight polymer solutions have better shear resistance and tend to equilibrate more quickly.
Currently, there are two main explanations for the mechanism: (1) Polymer interaction decomposition, polymer radicals react with dissolved oxygen to form polymer peroxyl radicals, and then Chain scission occurs by a bimolecular reaction between two polymer peroxyl radicals [7]; (2) Polymer chain decomposition, that is, mechanical breakage of the polymer chain. In our experiment, the 30 minutes of standing time as shown in Fig. 9 can not help the polymer long-chain molecules to recover their original drag-reducing ability, suggesting that the polymer molecular chains underwent mechanical breakage the chains. Consistent with the experimental findings of Zhang [42]. Meanwhile, Cussuol [33] found that after pre-shearing the PAM solution, there is a large decrease in the DR after 30 minutes, and even after 21 days only 83% of the initial value of the DR can be recovered, suggesting that irreversible breakage of the polymer chain occurred.
Mechanism discussion
The molecular structure of polymer long chains and the turbulent drag reduction behavior of polymer long chains are demonstrated in Fig. 10. The spatial structure, chain geometry, and flexibility presented by the polymer solution all have a large impact on the drag reduction behavior. Figure 11 shows the particle size distribution of PAM solution at different concentrations, it can be seen that the particle size of PAM in pure water is in the range of 164 nm at concentrations θ = 2ppm and 4ppm. Along with the increase in concentration, the particle size becomes larger and the range of the distribution becomes wider, which suggests that the increase in concentration results in the entanglement of the polymer long-chain molecules with each other. Similarly, the higher the molecular weight of the polymer at high concentration is, the deeper the curling and entanglement of the long-chain molecules are, and the lower the initial drag reduction is, as shown in Fig. 7.
Utilizing the intrinsic viscosity of the polymer, based on the Flory-Fox formula [43], the mean radius of gyration of the polymer can be determined.
$$[\eta ]={\Phi _0}{(\frac{{{h^2}}}{M})^{3/2}}{M^{1/2}}$$
6
$$R_{g}^{2}=\frac{{{h^2}}}{6}$$
7
Where \(\frac{{{h^2}}}{M}\)is a parameter characterizing the flexibility of the polymer chain; \([\eta ]\) is the intrinsic viscosity, mL/g; is the mean square end distance, cm; \({\Phi _0}\) is a universal constant independent of the nature of the polymer, \({\Phi _0}=2.84 \times {10^{23}}{\text{mo}}{{\text{l}}^{ - 1}}\);\(R_{g}^{{}}\) is the mean square radius of gyration, cm.
Since the polymer moves with the surrounding solvent and other particles as it flows in solution, the relaxation time of the linear flexible polymer is calculated using the Zimm theory [43, 44] and shown as shown in Eq. (8).
$$\lambda =\frac{{{\mu _S}R_{g}^{3}}}{{{k_B}T}}$$
8
Where kB is Boltzmann's constant, 1.38×10− 23 J·K− 1; T is the temperature, K; \({\mu _S}\) is the solvent viscosity, Pa.s.
Weissenberg number(Wi) involving relaxation time is defined as shown in Eq. (9).
$$Wi=\lambda \dot {r}=\lambda \frac{{8u}}{D}=\frac{{8u{\mu _S}R_{g}^{3}}}{{D{k_B}T}}$$
9
Where u is the average velocity, m/s; and D is the pipe diameter, m.
The Wi of PAM solutions of different molecular weights and concentrations are calculated using Eqs. (7) and (9) displayed in Table 3 and Fig. 12, from which the following conclusions have been obtained: (1) At the same concentration, along with the increase of Wi, the maximum drag reduction rate DRmax of different molecular weights increases, and the Wi number increases from 0.005 to 0.113, and corresponding to the increase of DRmax from 13.5–28.2%. Higher Wi implies that the solution has higher elasticity, which inhibits the vortex structure, suggesting that the elasticity of the polymer promotes drag reduction. The higher the elasticity is, the more obvious this facilitating effect is. (2) Comparing the Wi at different concentrations for a given molecular weight, it is found that an increase θ leads to an increase in the Wi corresponding to DRmax, suggesting that an increase θ promotes the elastic action of the polymer. (3) At DR < DRmax, the drag reduction rate increases with increasing Wi; whereas in the later stages of the experiment, the molecular weight of the polymer decreases due to the shear effect of the pump, which results in the curve starting to decrease after reaching DRmax.
Table 3
Maximum drag reduction DRmax and Wi for different molecular weights and concentrations
Molecular weights Mv(g/mol) | Rg (nm) | Concentration θ(ppm) | Maximum drag reduction DRmax(%) | Weissenberg number Wi |
663245 | 34.54 | 100 | 13.5 | 0.005 |
300 | 27.8 | 0.009 |
986454 | 43.03 | 100 | 17.0 | 0.015 |
300 | 29.4 | 0.027 |
1506913 | 54.40 | 100 | 18.4 | 0.032 |
300 | 31.3 | 0.061 |
2805193 | 76.72 | 100 | 28.2 | 0.113 |