APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Oct. 1985, p. 1112-1114 0099-2240/85/101112-03$02.

00/0 Copyright ) 1985, American Society for Microbiology

Vol. 50, No. 4

Passive Diffusion Technique for Concentration of Short-Chain Volatile Fatty Acids from Seawatert
JOHN J. MOLONGOSKI* AND CRAIG D. TAYLOR

Department of Biology, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543
Received 21 March 1985/Accepted 17 July 1985

A diffusion chamber is described which concentrates short-chain, volatile fatty acids from seawater while simultaneously separating them from interfering salts. The procedure relies on the passive diffusion of volatile compounds from acidified seawater samples and their subsequent absorption onto a base-impregnated filter. The method is simple, efficient, and adaptable to most commonly used methods of volatile acid analysis.

Despite the emergence of numerous techniques for the determination of volatile fatty acids (VFAs) (1, 3, 4, 6, 9), the quantification of these compounds in marine systems remains difficult (8). In part because of methodological and analytical constraints, measured concentrations of VFAs in various anoxic marine pore waters have been found to vary widely (i.e., 1.0 to 10,000 p.M for acetic acid) (2, 5, 7, 8). The detection of VFAs in saline environments is hindered both by their relatively low concentrations in these habitats and by the presence of large amounts of salt, which interferes with most analytical procedures. Vacuum distillation procedures have recently been used for the removal of salt from marine samples before VFA analysis (10, 11). Vacuum distillation has proved to be direct and efficient, but it is generally limited to small sample sizes and is not applicable to the rapid processing of many samples. We describe an alternative passive distillation procedure which concentrates volatile compounds such as acetic acid from seawater while simultaneously separating them from interfering salts. Our method relies on the passive diffusion of volatile compounds from an acidified seawater sample and their subsequent absorption onto a base-impregnated filter. The distillation is conducted in diffusion chambers (Fig. 1) consisting of a 1-in. (2.54-cm)-thick solid piece of polyvinyl chloride (4-in. [ca. 10 cm] outside diameter) which was machined to form a sample reservoir of large surface-tovolume ratio. Each chamber can hold up to 10 ml of sample. A sharpened center post (P) allows a potassium hydroxideimpregnated filter (F) to be suspended above the seawater sample. When the filter is in place, the sample is acidified to pH 1.0 or less, and the chamber is immediately sealed by clamping a 3/8-in. (ca. 0.95-cm) polyvinyl chloride lid (L) against a recessed 0 ring (OR) located around the chamber rim. Each chamber can be loaded such that the sample and acid do not mix until the chamber is sealed, preventing the loss of any volatile acid compound to the atmosphere. The sealed chambers are then stacked upright in multiples of four in stainless steel presses and incubated at 55C with gentle mixing on a rotary shaker (incubation temperatures exceeding 60C will cause softening or melting of the polyvinyl chloride chambers). During incubation, VFAs contained in the acidified seawater sample (S) diffuse out of solution (arrows in Fig. 1) and are absorbed onto the filter (F), where they are trapped as
*

their potassium salts. After diffusion is complete, the chambers are opened, and the filters are removed, dried at 60C, and stored desiccated over NaOH for subsequent VFA analysis. A time course of the volatilization and recovery of acetic acid by the passive distillation technique is shown in Fig. 2. [U-'4C]sodium acetate (2 ,uCi/ml) was added to seawater previously filtered through a Millipore-type GS membrane filter (0.22 p.m pore size) to yield a final concentration of radiolabeled acetate of 0.3 puM. Replicate 5-ml samples of the radiolabeled seawater were added to a series of diffusion chambers. Whatman 2.4-cm GF/F glass fiber filters, precombusted at 500C and then wetted with 250 ,ul of 0.5 N aqueous potassium hydroxide (equivalent to 125 ,umol of KOH), were used as trapping filters. Half of the seawater samples were amended with 250 p.l of absolute methanol; the other half received no methanol addition. Each sample was then acidified with 0.5 ml of 18 N H2SO4, sealed, and incubated as described above. Duplicate chambers were opened at the indicated times, and the filters and sample reservoirs were assayed for radioactivity by liquid scintillation spectrometry.

Corresponding author.
no.

t Woods Hole Oceanographic Institution contribution

5849.

The radiolabeled acetate contained in the methanolamended seawater samples was quantitatively recovered on the trapping filter after 12 h of incubation (KOH/MeOH filters; Fig. 2). Less than 1% of the added label remained in solution after 12 h (KOH/MeOH reservoirs). When methanol was not added to the seawater samples, the recovery of [14C]acetate on the filters was not complete even after 45 h (Fig. 2, KOH filters). Label not recovered on the filters remained in solution (Fig. 2, KOH reservoir). Thus, the addition of methanol to the seawater samples greatly accelerated the kinetics of acetate recovery on the filters. This enhancement was primarily the result of increased volatilization of acetate into the diffusion chamber atmosphere (Table 1). Methanol can be added directly to the seawater samples as in Fig. 2 or by wetting the trapping filter with methanolic rather than aqueous KOH. We found the two methods gave similar results (Table 1). The data in Table 1 also show the concentration of potassium hydroxide required for optimal recovery of acetic acid to be 0.5 N. Recovery of [14C]acetic acid from seawater by the passive distillation technique was constant over a range of six orders of magnitude of added acetate (Table 2). In addition, the procedure was applicable for the trapping of VFAs other than acetic acid. Nearly quantitative recovery was obtained for formic, propionic, and butyric acids (Table 3). Recovery
1112

VOL. 50, 1985

NOTES

1113

TABLE 1. Effect of methanol and potassium hydroxide concentration on the recovery of [14C]acetic acid by the passive diffusion techniquea ['4C]acetate (%)

OR

S.

KOH concn (N)

Trapped on filter

Remaining in solution

Recovered

C
FIG. 1. Apparatus for passive diffusion of volatile acid compounds from seawater samples.

of a less volatile compound, lactic acid, was not quantitative, however (Table 3). An experiment was also performed with nonvolatile ["4C]glucose as tracer to verify that the recovery of 14C on the trapping filter was the result of volatilizatiotl and diffusion and not of aerosol transport of sample onto the filter. As shown, essentially none of the ['4C]glucose (0.2% or less) appeared on the filters; rather, the label remained quantitatively in solution (Table 3). The passive diffusion technique is a simple and efficient method for desalting and concentrating VFAs from seawater. Once the chambers are constructed, the procedure is inexpensive to do and practical for use at sea, and it results in a space-conserving, easily storable product (i.e., desic-

0.10 (Methanolic filter) 0.25 (Methanolic filter) 0.40 (Methanolic filter) 0.50 (Methanolic filter) 0.50 (Aqueous filter) 0.50 (Aqueous filter; MeOH in sample) None None (MeOH in sample)

55.2 0.4
88.8 9.0

42.4 0.4 8.8 8.7 0.93 0.2 0.49 0.1 30.9 4.0
0.58 0.1

97.6 0.7 97.6 0.3 98.3 0.4 98.0 0.7 100.1 0.8
99.2 2.2

97.4 0.6 97.5 0.6 69.2 4.6 98.7 2.2 0 0

100.7 1.0b 91.6 + 3.5b.c

100.7 1.0 91.6 3.5

aThe percent recoveries reported are the mean values of duplicate determinations 1 standard error. Final concentration of [14C]acetic acid (2 ,uCi/ml) was 0.3 F.M. Samples (5 ml) were incubated with shaking for 24 h at 55C. bFurther incubation in either the absence or presence or MeOH resulted in no additional loss of [4C]acetic acid from solution, indicating that equilibrium between the sample reservoir and chamber headspace had been attained by
24 h. cLoss of radioactivity from solution was likely the consequence of enhancement of the equilibrium partitioning of [14C]acetic acid into the headspace by MeOH. Volatilized acetate was lost to the atmosphere when the chambers were opened.

TOTAL RECOVERY
100 80 100 75
50

I--w --_ _------cated glass fiber filters). Multiple samples can be processed simultaneously, depending on the number of chambers available. The time required for the recovery of VFAs can be reduced further if desired by raising the sample incubation temperature. As noted previously, however, that would necessitate constructing the chambers from a more heatstable material. The primary advantage of this technique is its potential for one-step sample concentration. Although the experiments described above used a 5-ml sample size, we quantitatively recovered 14C-labeled VFAs from 10-ml samples as well. The resulting filter can be resolvated in a minimum volume (i.e., 0.1 to 0.2 ml), resulting in a theoretical 50- to 100-fold

'b
1b..

25
0

100
75

50

TABLE 2. Effect of acetic acid concentration on recovery by the

passive diffusion techniquea

25

Acetate
0 0

[14C]acetate (%)
Trapped on
filter in solution

concn

16

24

32

40

48

(ALM)
10,000 1,000
100 10 1 0.30

RemainingReord
1.5 3.4 0.7 0.7 3.0 0.1

Recovered

T/ME(hours/ FIG. 2. Time course of volatilization and recovery of ['4C]acetic acid from seawater by passive diffusion technique. Final concentration of ["4C]acetic acid (2 ,uCi/ml) was 0.3 ,uM. A 5-ml sample of 'IC-labeled seawater was incubated with shaking for 24 h at 55C. Data were derived from two separate experiments: closed symbols,

96.4 1.7
98.2 96.9 97.7 97.5 97.4

0.2 0.5 0.3

2.9 0.4

3.3 3.0 1.1 0.51 2.7 0.55

99.7 101.2 98.0 98.2 97.2 98.0

0.2 3.1
0.2 0.4

0.07 0.3

experiment 1; open symbols, experiment 2. Symbols: * and O, total recovery, KOH trap alone; 0 and 0 total recovery with KOH trap plus seawater sample amended with methanol; 0 and 0, seawater samples not amended with methanol; A and A, seawater samples amended with methanol.
,

a The percent recoveries reported are the mean values of duplicate determinations _ 1 standard error. Final concentration of []4Cjacetic acid (2 .Ci/ml) added was 0.3 ,uM. Nonradioactive acetic acid was added to yield the final concentrations shown. Samples (5 ml) were incubated 24 h with shaking at 550C.

1114

NOTES

APPL. ENVIRON. MICROBIOL.

LITERATURE CITED 1. Ansbaek, J., and T. H. Blackburn. 1980. A method for the analysis of acetate turnover in a coastal marine sediment. Microb. Ecol. 5:253-264. 2. Barcelona, M. J. 1980. Dissolved organic carbon and volatile fatty acids in marine sediment pore waters. Geochim. Cosmochim. Acta 44:1977-1984. 3. Barcelona, M. J., H. M. Liljestrand, and J. J. Morgan. 1980. Determination of low molecular weight volatile fatty acids in aqueous samples. Anal. Chem. 52:321-325. 4. Bethge, P. O., and K. Liiedstrom. 1974. Determinaton of organic acids of low relative molecular mass (C1-C4) in dilute solution. Analyst (London) 99:137-142. 5. Christensen, D., and T. H. Blackburn. 1982. Turnover of '4Clabelled acetate in marine sediments. Mar. Biol. 71:113-119. 6. HordUk, K. A., and T. E. Cappenberg. 1983. Quantitative high-pressure liquid chromatography-fluorescence determination of some important lower fatty acids in lake sediments. Appl. Environ. Microbiol. 46:361-369. 7. Miller, D., C. M. Brown, T. H. Pearson, and S. 0. Stanley. 1979. Some biologically important low molecular weight organic acids in the sediments of Loch Eil. Mar. Biol. 50:375-383. 8. Parkes, R. J., and J. Taylor. 1983. Analysis of volatile fatty acids by ion-exclusion chromatography, with special reference to marine pore water. Mar. Biol. 77:113-118. 9. Sansone, F. J., and C. S. Martens. 1981. Determination of volatile fatty acid turnover rates in organic-rich marine sediments. Mar. Chem. 10:223-247. 10. Sorensen, J., D. Christensen, and B. B. J0rgensen. 1981. Volatile fatty acids and hydrogen as substrates for sulfate-reducing bacteria in anaerobic marine sediment. Appl. Environ. Microbiol. 42:5-11. 11. Tyler, J. E., and G. H. Dibdin. 1975. Gas chromatography of volatile fatty acids. Method involving separation from biological material by vacuum distillation. J. Chromatogr. 105:71-77.

TABLE 3. Recovery of "4C-compounds by the passive diffusion techniquea

4C-compound
Trapped on filter

14C (%) Remaining in solution

Recovered

Formic acid Acetic acid Propionic acid Butyric acid Lactic acid Glucose

96.2 97.4 98.2 95.9 20.4 0.2

0.95 0.4

0.28 1.8 4.6 0.01

0.7 0.55 0.93 2.7 83.0 100.4

0.2 0.1 0.53 1.3 t 2.1 1.07

96.9 98.0 99.1 98.6 103.4 100.6

0.75 0.3 0.21 0.50 2.5 1.08

a The percent recoveries reported are the mean values of duplicate determinations 1 standard error. Final concentration of each radiolabeled compound added (2 .CSi/ml) was 0.3 pLM. Samples (5 ml) were incubated 24 h with shaking at 55C.

sample concentration. Depending on the sensitivity of the analytical procedure used for the detection of VFAs (i.e., gas or high-pressure liquid chromatography), a lower limit of detectability of 1 ,uM or less of VFA can be obtained. Because the VFAs are concentrated on a glass fiber filter, the technique is easily adaptable to most commonly used methods of VFA determination (e.g., acidification and direct analysis by gas chromatography or esterification and detection by high-pressure liquid chromatography). The technique is now being used to measure microbially produced VFAs in anaerobic marine water columns.
This work was supported by national Science Foundation grants OCE79-19264 and OCE82-10420.