SUMMARY OF THE REMEDIATION TECHNOLOGIES DEVELOPMENT FORUM
PHYTOREMEDIATION OF ORGANICS ACTION TEAM
TOTAL PETROLEUM HYDROCARBON IN SOIL SUBGROUP
CONFERENCE CALL

May 29, 1998
10:30 a.m.—12:00 p.m.

On May 29, 1998, the following members of the Phytoremediation of Organics Action Team, Total Petroleum Hydrocarbon (TPH) in Soil Subgroup, met in a conference call:

Lucinda Jackson, Chevron Corporation (RTDF Action Team Co-Chair and Subgroup Co-Chair)
Phil Sayre, U.S. Environmental Protection Agency (EPA) (Subgroup Co-Chair)
Evelyn Drake, Exxon Research and Engineering
Jim Brown, Roy F. Weston
Jim Edwards, RETEC, Inc.
Peter Kulakow, Kansas State University (KSU)
Royal Nadeau, EPA, Environmental Response Team, New Jersey
Charlene Owens, Exxon Product Research
C.M. (Mike) Reynolds, U.S. Army Cold Regions
David Tsao, Amoco Research Center
Duane Wolf, University of Arkansas

Also present was Christine Hartnett of Eastern Research Group, Inc. (ERG).

COST ESTIMATES FOR THE HYDROCARBON ANALYSIS

Evelyn Drake opened the conference call by discussing cost estimates submitted by two laboratories: Arthur D. Little (ADL) and Batelle. Drake faxed the price quotes to conference call participants on May 28, 1998.

Drake said more precise cost breakdowns are presented in the fax that was sent to conference call participants. She said that she asked for detailed breakdowns because different teams within the RTDF Subgroup might choose to use the outside laboratories more extensively than others. For example, some teams may use the outside laboratory for all of the hydrocarbon analyses, while other teams may extract samples and perform TPH analysis in-house. Drake said that performing in-house analyses could be a good idea, but cautioned that using multiple laboratories increases the risk of generating inconsistent data.

Drake said that ADL and Batelle generated their cost estimates assuming that they would receive about 550 samples from the RTDF Subgroup over a 3-year period. Drake did not know whether the laboratories would raise their prices if they were to receive fewer samples.

Drake stressed that both laboratories are offering very reasonable rates. She said ADL's rate is about half what they charged during past projects. Batelle's price is even lower that ADL's reasonable rate. If the RTDF Subgroup decides to send all 550 samples to Batelle rather than ADL, the Subgroup will save about $50,000. Drake stressed that Batelle's price is very reasonable and thinks that they offered this rate because they really want the RTDF Subgroup's business. ADL is aware of Batelle's low rate but does not feel that they can match Batelle's price without losing money.

Drake said that she believes that both laboratories are competent and will offer equivalent capabilities. She said she has had extensive experience working with ADL and can vouch for their excellent quality of work. She also said that she was impressed with Batelle after talking with Batelle's Allan Uhler and Kevin McCarthy. In her discussions with Batelle, Drake was told that Batelle could complete analyses within 7 days for the initial site characterization sampling (T_I) and for the initial field test sampling (T₀) at no additional cost. (For all other sampling events, Drake asked for a 21-day turnaround.) Drake was confident that ADL would also be able to meet the RTDF Subgroup's turnaround times. Drake said that ADL has already agreed to store all the RTDF Subgroup's samples throughout the duration of the project and to perform all of the quality assurance/quality control activities that are required in normal government and industrial projects. Drake was confident that Batelle would offer to do the same.

Peter Kulakow asked whether the laboratories objected to the fact that they will receive samples in an intermittent fashion. Drake said the laboratories work most efficiently when they receive samples in batches of 12 to 20.

The following conference call participants expressed an opinion about which laboratory they prefer to use:

• Evelyn Drake. Drake would like to use the same laboratory as much as possible. Given Batelle's low price, she thought it would be wise to send the majority of samples to Batelle. She was reluctant, however, to give Batelle all the RTDF Subgroup's business. She and others (i.e., Lucinda Jackson and Royal Nadeau) praised ADL's Greg Douglas enthusiastically. As reported by Nadeau, Douglas has phenomenal experience, is very enthusiastic, and has an excellent reputation in the area of hydrocarbon analysis. Drake noted that she considers Douglas to be a long-time champion of petroleum analyses and does not wish to offend him by taking her business elsewhere. Drake said that she would like to send samples from the two Petroleum Environmental Research Forum (PERF) sites to ADL, since this laboratory performed the hydrocarbon analyses during the laboratory phase of the PERF project.

• Lucinda Jackson. Like Drake, Jackson said that she would also like the RTDF Subgroup to use the same laboratory as much as possible. She said she would be amenable to using either laboratory. Although her preference has always been ADL, she said she would be willing to use Batelle if other Subgroup members preferred Batelle.

• David Tsao. Tsao said that his decision will be driven mostly by economic considerations. He said he would be willing to use Batelle.

• Mike Reynolds. Reynolds said that his group has a contractual pathway established allowing them to use Batelle. He said it might be impossible for his group to use ADL.

• Phil Sayre. Jackson asked the conference call participants whether EPA has a preference of which laboratory to use. Sayre suspected that the Technology Innovation Office (TIO) would prefer to use Batelle because it is cheaper than ADL. Sayre stressed, however, that he is not sure of this.

• Duane Wolf. Wolf said that he will most likely be able to initiate activities at a site in Arkansas in the near future. At this point, he does not have a preference for which laboratory to use. He said he could probably use whichever laboratory the RTDF Subgroup decides is preferable.

• Peter Kulakow. Kulakow said that KSU plans to initiate activities at a field site soon. He is unsure, however, how much money will be available for outside laboratory contracts. He indicated that the bulk of the analyses may be performed in-house. Drake asked Kulakow whether KSU would consider sending an extract to one of the outside laboratories for confirmation of the in-house analyses. Kulakow thought this could be a possibility.

Given the opinions expressed by the conference call participants, Drake proposed having the RTDF Subgroup endorse both laboratories and then allowing individual project team leaders to choose which laboratory to use. Before deciding whether this is a viable option, however, she plans to gather additional information about whether (1) the laboratories are amenable to the idea of sharing the workload, and (2) the price per sample will increase if either laboratory receives less than 550 samples. Drake said that she will pursue these issues and will release a note summarizing her findings before the next RTDF Subgroup conference call. She said the note will also include information about using a standard sample to compare the results of both laboratories.

COORDINATING CENTER

Sayre said that he and Steve Rock have been talking about using a Hazardous Substances Research Center (HSRC) as a coordinating center that could help coordinate sample handling and data analysis. Sayre and Rock have been talking about using money contributed by TIO to partially fund the HSRC. Sayre noted that it is still unsure whether TIO will contribute money but that the probability is fairly high. Sayre suspected that the contribution could range from $20,000 to $60,000.

SAMPLE COLLECTION, SHIPPING PROTOCOL, AND SAMPLE STORAGE

RETEC, Inc. completed an SOP that describes how to collect, handle, store, and ship samples. Prior to the May 29, 1998, conference call, the SOP was distributed to the RTDF Subgroup participants. Jim Edwards, a RETEC, Inc. representative, summarized the SOP for conference call participants.

Edwards noted that the SOP recommends using a bucket auger and a split barrel core sampler to collect soil samples. (These devices can be ordered through forestry supply catalogs.) According to Edwards, these sampling tools help ensure that samples are collected from the desired depths. Edwards said that the SOP explains how to collect surface and subsurface samples for volatile organic compounds (VOCs), semivolatile organic compounds (SVOCs), TPH, and biomarker analyses. Edwards noted that the SOP describes a VOC sample collection methodology (SW-846 Method 5035) that has been adopted by some states (e.g., New Jersey) and will likely be adopted by additional states, such as New York, in the near future. This collection methodology is designed to prevent losses of VOCs due to volatilization and biodegradation prior to receipt at the laboratory.

Sayre questioned whether VOC samples need to be collected, stating that the majority of the RTDF Subgroup's proposed sites have fairly aged hydrocarbons. Tsao said that VOCs are not expected to be prevalent at his site. Jackson said that she will collect VOCs at the Chevron site in Ohio simply because EPA is requiring her to do so. She thanked Edwards for providing information on the SW-846 Method 5035 methodology because EPA has asked her to use this methodology. Reynolds said that he plans to collect VOC samples at T₀ and 30 months after planting (T₃). Both Jackson and Tsao agreed that sampling at the beginning and end of the field test could help address regulatory questions.

Jackson remarked that the SOP does not stress the importance of icing samples in the field. Edwards said that the SOP's "Packaging and Shipment of Samples" section states that samples should be iced, but he acknowledged that the SOP fails to indicate that the samples should be iced immediately upon collection.

Sayre said that Steve Geiger, also from RETEC, Inc., had indicated that it would be good to include benzene and carcinogenic PAHs in the SOP. Sayre asked Edwards whether these issues were addressed in the SOP. Edwards said he would check with Geiger and get back to Sayre.

As pointed out by Sayre, RETEC Inc.'s SOP states that soil samples should be collected in glass jars even though Drake recommended against using glass jars in a previous conference call. Edwards and Jackson said that they have always used glass jars throughout their careers.

Conference call participants said that the SOP was concise, easy to understand, and extremely detailed. Reynolds noted that it is wise to have a detailed plan before going into the field. The participants agreed that the SOP should be added as an attachment to the protocol. Kulakow asked how closely RTDF Subgroup participants would be expected to follow the protocol. For example, he asked, would a team member be breaking protocol if they decided not to test for VOCs? Jackson and Edwards noted that VOC sampling is listed as optional in the SOP.

SEED MIX AND PLANTING DENSITY

Kulakow completed an example of a generalized grass seeding mixture design for soil contaminated with petroleum hydrocarbons. Prior to the May 29, 1998, conference call, Kulakow's design example was distributed to conference call participants. Kulakow summarized the main points:

Selecting a seed mixture and a planting protocol needs to be site-specific. Kulakow said that it is important for investigators to understand local environmental conditions, soil properties, and contaminant characteristics at a site before determing what plants to select. He said that investigators can gather site-specific information by consulting with local agricultural extensions, the Natural Resource Conservation Service, and local seed suppliers.

• Phytotoxicity can limit seed germination and plant growth rates at sites that have high TPH levels. Kulakow said that he has not found phytotoxicity to be a large problem at sites where TPH levels are less than 5,000 parts per million (ppm). At levels of 40,000 ppm, he said, investigators can nearly be guaranteed to encounter phytotoxicity problems. Kulakow stressed that many plants can still grow in highly contaminated areas but that their germination rates are greatly reduced. Kulakow recommended performing germination trials at sites where concentrations are high. Kulakow said these trials can be performed in the field, in the greenhouse, or in the laboratory (i.e., in petri dishes). He said that investigators can choose higher seeding rates to compensate for the reduced proliferation that plants exhibit in highly contaminated areas.

• Legumes should be inoculated with the appropriate Rhizobium strain to facilitate effective nitrogen fixation. Kulakow said he has seen highly effective nodules form at some of the sites he has worked at. Kulakow said that investigators can either (1) purchase legumes that have already been inoculated, or (2) inoculate legumes by mixing a bag of legume seeds with a bag of inoculum. Jim Brown told participants that they should ask seed suppliers if the legume seed coats have been scarified before purchasing the seeds. (Seeds need to be "roughed up" to promote germination.)

• The relationship between initial seeding and overseeding (i.e., reseeding) rates should be site-specific. Jackson noted that Kulakow's example suggests that initial and overseeding rates should be the same. Kulakow said his experience has shown this approach is useful. He stressed, however, that he would choose to double the initial rate if the initial seeding establishment was poor. He emphasized that it is important to order enough seed before initiating activities and to be prepared to overseed. In his example, Kulakow recommended using 8 pounds of seed per 1,000 square feet for tall fescue, 2 pounds of seed per 1,000 square feet for annual ryegrass, and 1 pound per 1,000 square feet for legumes. He stressed that these are approximate figures and should not be used as a guideline for all sites.

Kulakow told the conference call participants that they should feel free to call him if they want to discusss site-specific issues.

SOIL FERTILITY EVALUATION

Brown recommended collecting a composite sample during T_I and sending it to a regional soil testing laboratory for a soil fertility evaluation. This evaluation, he said, would provide a variety of useful information, including native nutrient concentrations and pH levels. Brown estimated that the analyses would cost about $7 to $8 per sample. Brown encouraged RTDF Subgroup participants to establish a relationship with local agricultural extension personnel and to inform them of the Subgroup's activities.

Reynolds asked whether standard soil testing procedures are valid for samples that are oily. Brown said certain tests, such as the cation exchange capacity test, can be rendered ineffective if the samples are too oily. To avoid problems, Brown said that he would collect nearby uncontaminated soil samples if the contaminated soils are very oily. In general, Brown did not think problems would be encountered unless the sample was greater than 2 to 3% (20,000 to 30,000 ppm) oil.

Fertilization Rates

Brown emphasized the importance of using fertilizer at test sites. Without fertilizer, he posited, a competition for nutrients could be established between microbes and plants. His experience has shown that such competitive scenarios lead to unhealthy, nutrient-deficient plants. Brown said that many of the former land treatment test sites will likely have high nutrient loads, but stressed that investigators should ensure that this is the case. Any deficiencies in nitrogen or phosphorus, he stressed, could have adverse effects on phytoremediation field studies. As a general rule of thumb, Brown suspected that Subgroup participants would need to add about 100 pounds of nitrogen per acre and 25 to 30 pounds of phosphorus per acre. He stressed, however, that much higher loads may be required if areas are nutrient deficient. Charlene Owens said that sites with a high hydrocarbon content will need large fertilizer quantities to support the microbial degrader population. She warned that fertilizer quantities could be high enough to cause toxicity to plants. She recommended, therefore, that fertilizer be added intermittently rather than all at once. Brown agreed with Owens' point, noting that adding excessive nitrogen can cause adverse salt effects, toxicity, and water quality problems. Kulakow said that his group typically adds fertilizer to test sites every 3 to 6 months.

Reynolds noted that the current protocol calls for an unfertilized control plot and fertilized plant treatment plots. He pointed out that none of the plots will have just plants or just fertilizer. If phytoremedial effects are observed, he noted, it could be difficult to determine whether it is the fertilizer or the plants that are causing the phytoremediation, particularly if one research team opts to use a far greater proportion of fertilizer than another. Owens asked the group whether it would be wise to include a fertilized control plot to answer some of these questions. Jackson said that she likes the protocol as is because it is focused and simple. Reynolds recommended trying to identify a fertilizer application scheme that could be adopted by the group. For example, he said, maybe all of the teams should determine the rate of fertilizer that is required at their sites and then double that rate. Sayre recommended discussing this topic in more detail during the next conference call.

Lime

Brown noted that the activities of hydrocarbon degraders often create acidic conditions. He recommended testing soil pH to ensure that soils are not overly acidic. If the soil is in the mid-5 range, he said he would add lime until the soil pH reaches the higher end of the 6 range. Jackson agreed, saying that this was an excellent point and that she would change the protocol to reflect Brown's comment.

MICROBIAL ANALYSIS PROTOCOL

Deciding Whether It Is Necessary to Perform Microbial Analysis

Jackson asked the group whether it is truly necessary to perform microbial analyses. She noted that her clients are interested in the end results of phytoremedial technologies and are less concerned with determining why the technology works. She asked whether the microbial analysis could be considered an optional analysis.

Sayre noted that the protocol was originally designed to have a tiered approach, clearly identifying which analyses are considered essential and which are optional. Sayre said that the tiered approach could be reinstated, but Jackson said that she likes the protocol as it is because it is simple and straightforward.

Brown said that the microbial analysis should be optional. In cases where limited funds are available, he believes investigators would benefit more strongly by focusing their efforts on other analyses. Compared to other analyses, Brown said, the cost-benefit relationship for microbial analysis is in the lower third.

Reynolds said that the microbial analysis should be considered optional because, in reality, all of the analyses are optional and will only be performed if adequate funding becomes available. Reynolds strongly encouraged participants to perform microbial analysis, however, and stressed that the RTDF Subgroup will miss a golden opportunity to determine how phytoremediation works if they fail to do so. Reynolds said that the RTDF Subgroup's protocol is unique because it includes a control plot in the test design. In the future, Reynolds doubted that investigators will be as willing to include control plots.

Tsao said that collecting microbial analysis samples is important because some state regulators may require proof of how phytoremediation works.

Jackson said the conference call participants convinced her that collecting the samples is important and she said she will make an effort to collect samples for microbial analysis.

Deciding When to Collect Samples for Microbial Analysis

Conference call participants debated whether initial microbial analysis should be performed during the T_I or T₀sampling event. Sayre said that the T_I sampling event is designed to assist in picking actual plots within a test site. He argued, therefore, that microbial analysis samples should be collected at T_I to ensure that the selected plots support a sufficient number of hydrocarbon-degrading microorganisms. Sayre warned that investigators run the risk of establishing plots in doomed areas if they forego the T_I sampling. Initially, Brown supported Sayre's argument. Later in the conversation, however, Brown detracted from the argument by noting that investigators can feel reasonably confident that hydrocarbon degraders will be present in soils since they are so ubiquitous in the environment. Brown said that most of the RTDF Subgroup's proposed test sites have fairly weathered soil that likely foster a robust degrader population. Even at recent spill sites, he commented, degraders will likely be present because hydrocarbon degraders are typically present in aerobic environments. Brown stressed that investigators may only need to sample at T_I if their purpose is to establish plots that have a high number of a specific type of degrader (i.e., degraders that attack the recalcitrant carcinogenic PAHs.)

Jackson said she thinks it would be best to collect a couple of samples at T_I to get a casual look at the microorganism distribution but recommended postponing a more detailed analysis until T₀. She asked participants whether the microbial analysis samples should be collected (1) plot by plot (i.e., collect 4 replicates per plot) so that statistical differences can be calculated or (2) as a composite so that investigators can get a general idea of the conditions across the treatment areas.

Brown recommended collecting composite samples for the microbial analysis sample. He said that the variability in the microbe data might be so great that a huge number of samples would be required before any statistical differences could be shown across treatments.

Wolf said that his group plans to do a plot-by-plot analysis. Wolf said previous work indicates that statistically significant 1,000-fold differences can be detected between treatments. Wolf's group plans to perform microbial analyses on multiple replicates during the T₀ sampling event. By doing so, they will be able to detect whether different plots supported different numbers of microorganisms before the phytoremedial experiment was initiated. If differences are not detected at T₀ but are detected at T₃, he noted, his group will have strong evidence indicating that the rhizosphere effect is important.

Conference call participants agreed that microbial analysis samples should be collected at T₃. Wolf and Reynolds said that they will collect microbial samples a couple of times before T₃ as well. Reynolds noted that his sample collection times will be dictated by the weather rather than specific pre-selected time periods. Because his sites are in Alaska, Reynolds will take care to collect samples when the soil is not frozen.

Oil Extraction Technique

During the May 19, 1998, conference call, the RTDF Subgroup decided that they wanted to use a site-specific oil for their bacterial enumeration analysis. The Subgroup was generally satisfied with an extraction methodology that was provided by Brown, but Reynolds had asked whether significant amounts of highly decomposed organic matter carbon would be extracted along with the hydrocarbons, and if so, how that would effect the enumeration analyses.

During the May 29, 1998, conference call, Brown tried to answer Reynolds question. Brown admitted that he did not have a solid answer but said that he consulted a Micro-Innotech representative and asked him whether extracting native, highly decomposed soil organics along with hydrocarbons could have a significant impact on the enumeration analysis. The representative was unwilling to offer a definite answer but said that he doubted that the impact would be measurable. Brown agreed with the representative's opinion.

Collecting and Handling Microbial Analysis Samples

Brown warned that the quality of outgoing laboratory results hinges upon the quality of incoming samples. He stressed the importance of cooling samples and maintaining field moisture content. Failure to do so, he said, can cause microorganisms to die from stress. Brown said that he allowed some samples to air dry a bit before sending them for analysis. When he received bacterial counts from the laboratory, total heterotroph counts were two orders of magnitude lower than he expected (i.e., 10⁴ per gram rather than 10⁶ or 10⁷). Edwards said that RETEC, Inc.'s SOP is designed to keep samples cool and at their field moisture levels.

Choosing a Laboratory to Perform Microbial Analyses

Owens and other participants said that it is not crucial for all microbial analysis samples to be processed by the same laboratory. Many teams within the RTDF Subgroup may choose to perform the analyses in-house. The Subgroup has identified an oil extraction technique and a protocol for growing and analyzing the microorganisms to help ensure that the individual teams remain consistent with one another. The Subgroup has also identified a laboratory, Micro-Innotech, that can perform the analysis if certain teams prefer to use an outside laboratory.

MISCELLANEOUS

Sayre asked Edwards whether RETEC, Inc., has a potential test site. Edwards said that Sayre should discuss this with Joe Kreitinger.

ACTION ITEMS

• Drake said that she will distribute a note to conference call participants before the next call. The note will include additional information about the possibility of splitting the sample analysis workload between ADL and Batelle. The note will also include information about using a standard sample to compare the results of both laboratories.

• Jackson and Kulakow agreed to collaborate and to produce a protocol for plant assessment (i.e., assessing percent cover and shoot height) and for plant sampling.

• Sayre said he would ask Rock about the possibility of including a statistician on the next conference call.

NEXT CONFERENCE CALL

The next conference call is scheduled for June 5, 1998, between 12:00 p.m. and 1:30 p.m. (EST).

REMEDIATION TECHNOLOGIES DEVELOPMENT FORUM
PHYTOREMEDIATION ACTION TEAM
FIELD STUDY PROTOCOL
Steve Rock's Version