Identification of Volatile and Semivolatile Organic Compounds in Salt Lake Brine

The deposited salts and brine from salt lakes are important industrial resources for many kinds of salts, such as edible salt, potassium, boron, lithium, and a few heavy metals. Nearly 97% of potassium fertilizer made in China is produced from the salt lake brines. However, in the industrial process for manufacturing the targeted products from salt lake brines, organic substances in brine have been found to influence the quality of products and the manufacturing process. An effective sample preparation and analysis method is required in order to determine the trace organics in brine, including volatile organic compounds (VOCs) and semi-volatile organic compounds (SVOCs).

The conventional methods of screening organics are liquid–liquid extraction, solid-phase extraction (SPE),1,2 solid-phase microextraction,3 purge and trap,4 reverse osmosis, and electrodialysis.5 Brown6 reported that XAD-8 and XAD-4 could be used to absorb dissolved organic material for isolating fuvic acids and transphilic acids from Pony Lake, a saline and hypereutrophic coastal pond located on Cape Royds in the McMurdo Sound area in Antarctica. GC-MS is typically used to determine VOCs and SVOCs. In the present study, routine analyses were performed on a QP2010 GC-MS (Shimadzu Corp., Kyoto, Japan).

Brines in the Qarhan Salt Lake (Haixi, China) are hypersaline water and usually saturated by salts. XAD mixed resins and LC-C18 cartridges were used to collect and concentrate the organics in the authors’ studies. The analytes were identified by GC-MS.


Site description

Qarhan Salt Lake, located in south central Qaidam basin between 94°28’ and 96°23’ north latitude and 36°43’ and 37°02’ east longitude, is the largest playa in China. The lake is about 168 km in length and 20–40 km in width. Typically, the brine is neutral, with an average salinity of 249.5 g L–1 and a specific gravity of 1.201.7

In this study, water samples were collected from the carnallite pond of a potassium fertilizer plant (Qinghai Salt Lake Industry Group Co., Ltd., the largest potassium fertilizer industry in China). The brines of the pond are from Qarhan Salt Lake and are concentrated by sunshine.

Reagents and materials

Naphthalene-D10 internal reference was supplied by Acros Organics (Morris Plains, NJ) with 98+ atom% deuterium. Working standard solutions were a series of ethyl acetate solutions containing naphthalene-D10 between 0.1 mg L–1 and 1 mg L–1. The standard solutions were freshly prepared and were stored in the refrigerator at 4 °C.

Methanol, acetone, methylene chloride (DCM), and ethyl acetate (spectrograde) were further distilled before use. Magnesium sulfate (analytical grade) was heated in a 300 °C muffle furnace for 3 hr to remove organic substances and water before use. Water used throughout this work was freshly double distilled from potassium permanganate (ddH2O).

LC-C18 solid-phase extraction cartridges (Supelco, Bellefonte, PA) and XAD-2 (Rohm and Haas, Philadelphia, PA) and XAD-7 resins (Acros Organics) were used for solid-phase extraction. The properties of the absorbents are presented in Table 1. The resins were purified by sequential solvent extraction with methanol, DCM, and acetone in a Soxhlet extractor (8 hr per solvent). The purified resins were stored in glass stoppered bottles under methanol to maintain their high purity.

Table 1    -    Properties of absorbents


The solid-phase extraction equipment was supplied by Sigma-Aldrich Corp. (St. Louis, MO). A KL 512 nitrogen evaporator (Kanglin Science & Technology Co., Ltd., Beijing, China) was used for concentration.

As stated above, routine analyses were performed on the QP2010 GC-MS. A DB-5 MS capillary column (Agilent Technologies, Palo Alto, CA) (30 m × 0.25 mm × 0.25 μm film thickness) was employed for the separation of the analytes. GC conditions were as follows: injector temperature, 240 °C; flame ionization detector temperature, 240 °C, split 1:20; oven temperature, 50 °C, then 8 °C min–1 to 200 °C (hold 5 min), then 10 °C min–1 to 240 °C (hold for 5 min). MS conditions were as follows: electron impact ionization, ion energy to 70 eV; mass range, 35–350 m/z.

Sample preparation

The original brines are usually saturated with salts. If the original brines were passed through the LC-C18 cartridge and XAD column directly, salts might be precipitated, which causes the jam of columns. To avoid jamming the LC-C18 cartridge and XAD column and to maintain a moderate flow rate in the pretreatment procedure, 500 mL brine sample was diluted with ddH2O to 1 L. The concentration of the internal standard in the diluted sample was 0.5 µg L–1.

C18 cartridge solid-phase extraction procedures were as follows. The diluted sample was passed through the LC-C18 cartridge previously conditioned by passing methanol (10 mL), ethyl acetate (8 mL), and ddH2O (10 mL) in sequence. After loading the sample, the cartridge was washed with ddH2O (10 mL). The cartridge was air-dried using vacuum for at least 15 min, and then eluted with 10 mL ethyl acetate and 10 mL DCM. The proper amount of anhydrous magnesium sulfate was added to the extracted solution to remove the residual water until a clear solution was obtained. After filtering, the collected solution was then evaporated to 1 mL under a gentle nitrogen stream at 50 °C. The final extract obtained was analyzed and identified by GC-MS.

The procedures reported by Junk8 were referred to for the XAD resin solid-phase extraction. A 1.5-cm-i.d. × 20-cm-long glass tube was packed with XAD-2 and XAD-7 (v/v 4:1). The resin was then washed with ddH2O to remove methanol. The diluted sample was passed through the XAD resin column by gravity flow. After loading the sample, the column was washed with 20 mL of ddH2O. Subsequently, the column was dried by passing air, and the elutions were performed by passing 15 mL ethyl acetate and 15 mL DCM in sequence (each elution solvent was allowed to equilibrate with the resin for 10 min). Finally, an additional 5 mL of DCM was added to the column and immediately flowed through the resin. The residual water was removed by adding the proper amount of anhydrous magnesium sulfate to the extracted solution until the solution became clear. After filtering, the collected extract was evaporated to 1 mL under a gentle nitrogen stream at 50 °C. The concentrated solution was analyzed and identified by GC-MS.

Results and discussion

Solid-phase extraction

The authors selected naphthalene-D10 as the internal reference for the recovery experiments. The calibration curves were obtained by analyzing a series of naphthalene-D10 standard solutions with a concentration between 0.1 and 1 mg L–1.

Table 2    -    Recovery and standard deviations (SDs) of the method

The concentration of naphthalene-D10 in the diluted sample was 0.5 μg L–1. The measured results are given as recoveries in Table 2. The mean recovery of the LC-C18 cartridge was higher than the XAD column. The low recovery of the XAD column may be due to the air bubbles formed in the process of solid-phase extraction. These bubbles may have caused a bypass effect that reduced the efficiency. This phenomenon was not be found in C18 solid-phase extraction because the sorbent in the LC-C18 cartridges was tightly packed. It can be argued that the different affinities between LC-C18 and XAD resins to naphthalene-D10 can also affect the efficiency.