TOWPF was taken from an offshore oilfield in china. In this offshore oilfield, oily wastewater treatment process is shown in Figure 1.

The oil displacement agent was hydrophobic modified polyacryamide (HMPAM). Therefore, there was residual HMPAM in the produced wastewater. The produced wastewater containing HMPAM was treated by skimmer, air floatation and walnut shell filter in turn and TOWPF was obtained [10] and came to buffer vessel. Its quality is listed in Table 1. In general, the TOWPF would be stay in buffer vessel for about 2 - 48 h. However, during this period, precipitates containing HMPAM would appear at the bottom of buffer vessel [11] [12]. This phenomenon represented that the stability of residual HMPAM in TOWPF was destroyed by some reason.

Pyrene was received from Aladdin-Reagent Corporation (Shanghai, China).

Fluorescence spectrum was measured by LS55 fluorescence spectrophotometer (PerkinElmer Corporation, England).

Dynamic light scattering (DLS) experiments were performed using Brookhaven Instruments 90 Plus/BI-MAS (Brookhaven Instruments, NY, USA). Hydrodynamic radius (R_h) was investigated by the DLS and calculated according to the CONTIN algorithm.

Size exclusion chromatography (SEC) analysis was conducted using a Water 515 liquid chromatograph instrument connected with a Waters 2410 refractive index detector at 30˚C. The gel permeation column was linear ultrahydrogel (7.8 × 300 mm²), and the

solvent used was distilled water. Polymer standards of Dextran were used from NICPBP (National Institute for the Control of Pharmaceutical and Biological Products). The samples were dissolved in 0.1 M NaCl solution and analyzed at a flow rate of 0.8 mL∙min⁻¹. Number-average molecular weight (M_n), weight-average molecular weight (M_w) and polydispersity index (PDI) were measured by SEC.

Intrinsic viscosities ([η]) of polymers were measured with the “five-spot”dilution method using an Ubbelohde viscometer at 30˚C, where the solvent was 1 M NaCl solution. Viscosity-average molecular weight (M_w) was calculated as follows [13]:

Firstly, 2 L TOWPF was concentrated to 200 mL. Then, the concentrated TOWPF was purified by osmotic membrane with molecular weight cut off 10,000 for three days. At last, after vacuum drying for 8 h under the condition of 60˚C, the HMPAM in TOWPF was obtained. The structure of HMPAM is shown in Figure 2.

The characterization of residual HMPAM is listed in Table 2. It can be found that its molecular weight was not high, but its polydispersity was high. Because its molecular weight was not high, its mobility may be similar with the small molecule surfactant. There were intermolecular aggregation for residual HMPAM.

Figure 4 shows the fluorescence spectrums of blank residual HMPAM solution at different layers and different times. The fluorescence spectrums of blank residual HMPAM solution at different layers were very close at the same time. As time increasing, the fluorescence intensity increased. Table 3 lists the I₁/I₃ of blank residual HMPAM solution at different layers and I_1,t/I_1.0h at different times. The results showed that the I₁/I₃ of blank residual HMPAM solution at different layers were closed and were about 0.85. Interestingly, the I₁/I₃ was not influenced by time. This phenomenon presented that the polarities of hydrophobic domain formed by residual HMPAM at different layers of blank residual HMPAM solution were the same. Therefore, the blank residual HMPAM solution was homogeneous and the polarity of hydrophobic domain was not changed during 48 h. Meanwhile, the increasing I_1,t/I_1.0h represented that the number of hydrophobic domain could increase with time because the formation of hydrophobic domain was a slow process for blank residual HMPAM solution.

Table 4 shows the influence of Fe²⁺ on the fluorescence spectrums of residual HMPAM solution at different layers and different times. After the addition of Fe²⁺ into residual HMPAM solution, the I₁/I₃ was much larger than that of blank and not influenced by time too. It showed that the polarity of hydrophobic domain was increased. The reason of Fe²⁺ for increasing I₁/I₃ may be the chelation between these substances and carboxyl group in HMPAM. After the chelation, the moving ability and rotation ability of HMPAM decreased and the hydrophobic group used to formation of hydrophobic domain was decreased (see Figure 5). Therefore, the polarity of hydrophobic domain increased.

In addition, as shown in Table 4, after the addition of Fe²⁺, the I_1,t /I_1,0h increased with time,. Compared to the I_1.48h/I_1.0h of blank HMPAM solution. It can be found that the addition of Fe²⁺, the HAPAM solution had much larger I_1.48h/I_1.0h than that of blank. It represented that the Fe²⁺ was helpful for the increase of hydrophobic domain. Usually, the increasing hydrophobic domain means the increasing aggregation degree. In summary, when there was Fe²⁺, HMPAM solution had the largest aggregation degree (I_1.48h/I_1.0h = 1.89); After the addition of Fe²⁺, the chelation between Fe²⁺ and carboxy group in HMPAM would increase with time during 48 h. Therefore, the number of hydrophobic domains increased with time obviously.

Usually, the increasing hydrophobic domain may be caused by increasing number of HMPAM aggregates or increasing size of HMPAM aggregates. The results of DLS can reflect that whether the size of HMPAM aggregate increased or not. Figure 6 and Table 5 showed the effects of Fe²⁺ on the relationship between R_h of HMPAM at different

time. For blank residual HMPAM solution, the R_h had no great change with time. When there was Fe²⁺ in solution, the R_h had no great change with time, too. The results showed that the increasing of I_1,t/I_1.0h was caused by the increasing number of HMPAM aggregate.