785-825-1842 001-785-825-1842 785-825-1842 001-785-825-1842

LL MIP

Low Level Membrane Interface Probe


Lower your MIP detection limits and increase your system usage with the LL MIP Pulse Flow controller

LL MIP was developed as a way to reduce MIP detection limits and offers the capability of reaching well below 100ppb total VOCs. These lowered detection limits means operators equipment sees more days in the field as plume investigations go to further extents and sites that MIP equipment would previously not have been used on are being investigated.

LL MIP Pulse Flow Controller

System Overview



Low Level MIP (LL MIP) is a technology developed by Geoprobe Systems® that greatly increases the sensitivity (and therefore utility) of the MIP logging tool.  The primary feature of LL MIP technology is that the carrier gas stream that sweeps the internal surface of the MIP membrane is pulsed.  This results in an increase in the concentration of VOC contaminant delivered to the MIP detectors.

Low Level MIP can be performed with any style of MIP probes or trunkline.  To perform this method an operator will need to add the MP9000 Pulse Flow Controller (Figure 1) to their current MIP system and be using a current version of DI Acquisition software.  The addition of the MP9000 to the system is simple and requires only the rearrangement of gas line connections.  This controller can then be easily removed from the system to return to standard MIP logging.  Switching between methods requires only a few minutes of time.

LL MIP Equipment: GC and detectors, FI6000, MP9000, MP6505

Figure 1: LL MIP Equipment: GC and detectors, FI6000, MP9000, MP6505

General Operation:
In standard MIP operation, the carrier gas continually sweeps across the membrane transporting contaminates to the detectors at the surface.  In the LL MIP method, the trunkline sweep flow is temporarily stopped when the MIP probe is brought to rest at a discrete depth in the soil.  Stopping the sweep gas flow allows the contaminant concentration to build behind the membrane.  This results in a larger and narrower contaminant response peak at the detectors (Fig. 2 and 3) for a given chemical concentration.  Switching valves located inside the MP9000 create separate flow paths for the MIP trunkline and detectors; trunkline flow can be stopped and restarted without impacting detector baseline or stability.   When the trunkline flow is restarted the contaminant mass (peak) is quickly swept to the surface with a trunkline flow rate of approximately 60ml/min. and is routed to the detectors via the MP9000.

Comparison of 0.5ppm Trichloroethylene response between standard MIP (50s-100s) and Low Level MIP (300s) methods

Figure 2: Comparison of 0.5ppm Trichloroethylene response between standard MIP (50s-100s) and Low Level MIP (300s) methods

The Low Level MIP operation is automated in DI Acquisition software version 3.0 or newer which controls the timed cycling of the MP9000 controller.  This software includes a subroutine for automatic triggering of LL MIP cycling at preset depth intervals during the MIP logging process.
The LL MIP method will greatly increase the sensitivity of a MIP system but the resulting detection limits are dependent on the sensitivity of the standard MIP system.  To achieve the lowest possible detection limits the probe and trunkline need to be new or verified clean with system blanks.  The detectors also need to be fully current within their maintenance program and sensitivity should be tested prior to mobilization.  Equipment that has been used to map high level contaminants will result in false positive results due to contaminant desorption from the membrane and return carrier gas line. 

Comparison of 0.2, 1.0, and 5.0ppm Trichloroethylene in standard and LL MIP methods

Figure 3: Comparison of 0.2, 1.0, and 5.0ppm Trichloroethylene in standard and LL MIP methods

Example Logs

The LL MIHPT Log in Figure 4 shows a highly permeable likely sandy formation in the top 22ft as indicated by the low EC and HPT pressure values.  Permeability stays fairly low throughout the log with occasional layers of lower permeability.  The predominant contaminant is seen from approximately 37ft to 51ft and is detected primarily on the PID and XSD.  Since the compounds respond on the XSD we know them to contain halogenated probably double bonded halogenated due to the PID confirmation, which to respond on the PID requires an ionization potential below the lamps 10.6eV.  The halogenated compounds that have lower ionization potential are typically double bonded chemicals.   The FID shows some responses within the first 5ft of the log which the other detectors do not confirm.  This is a common trait of natural gas or methane deposits which is non-halogenated and has a higher ionization potential than the lamp.  In the lower contaminant zone there is a confirmed response on the FID although lower in magnitude than the PID and XSD due to the lower sensitivity of the FID for these compounds.  Another item to note is that the PID displays consistent responses throughout the log, if it quite common for the PID to provide a background level response in LL MIP mode and the lower responses must be disregarded.

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, XSD, FID, and Est. K.

Figure 4: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, XSD, FID, and Est. K.

The LL MIHPT Log in Figure 5 shows a lowery permeability soils such as silts and clays in the top 45ft as indicated by the higher EC and HPT pressure values.  This site breaks into a sandy formation below 45ft indicated by the lower EC and HPT pressure below this point.  The predominant contaminant is seen from approximately 15ft to 33ft and is detected primarily on the PID and FID with no XSD response.  Since there is no response on the XSD we know it is a non-halogenated compound.  This log came from a former fuel service station so the PID and FID response likely is aromatic and nonaromatic fuel hydrocarbons.  

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

Figure 5: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

The LL MIHPT Log in Figure 6 shows a lower permeable clay at the top 5ft transitioning into likely a sandy silt formation down to 30ft as indicated by the low EC and HPT pressure values.  Below 30ft the formation is a fairly clean sand with low EC and HPT values. Permeability stays low throughout the log with occasional layers of lower permeability below 50ft.  There are different contaminants in this log with likely petroleum hydrocarbons in the top 30ft responding on the PID and FID only but then the XSD shows halogenated compounds 35ft-42ft and again from 52ft to the end of the boring.  These XSD responses are not confirmed on the PID which indicates that the ionization potential is likely too high for the PID to detect.  This typically occurs with single bonded halogenated compounds.  Confirmation groundwater samples were taken from these locations and 1,2-Dichloroethane was the predominant contaminant in the zone from 35ft to 42ft, in the lower section it was mostly Chloroform and Carbon Tetrachloride.   All of these compounds have ionization potentials over 11eV, so they would not be detected by the PID.   

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

Figure 6: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

MIP Method Comparison

Low level MIP will not replace standard MIP logging, but will expand MIP capabilities.  LL MIP is most useful when low contaminant concentrations are present.  Figure 7 displays an overlay of closely spaced standard MIP and LL MIP logs at locations of lower concentration petroleum and chlorinated VOC’s.  The LL MIP log exhibits robust signal to noise and the presence and location of contamination is easily defined.  The standard MIP log you might be able to discern there is a contaminant if you zoom in on the baseline but at times you can wonder if the basleine  movement is due to contaminant or is detector baseline drift.  This log displays petroleum hydrocarbon presence in the top half of the log with PID-FID responses and no XSD response and in the lower half of the log there are chlorinated contamoinants present as displayed by the PID-XSD responses.  Projects with contaminant concentrations too low to proceed with traditional MIP methods may be good candidates for investigation using the LL MIP method.  This technology will provide operators with the ability to track and map contaminant plumes down to concentrations at or below the 100ppb range for some contaminants.

Collocated standard (shaded green) and LL MIP Logs overlaid displaying (left to right) EC, MIP-PID, FID, and XSD

Figure 7: Collocated standard (shaded green) and LL MIP Logs overlaid displaying (left to right) EC, MIP-PID, FID, and XSD

The MIP logs compared in Figure 8 display closely spaced standard and LL MIP logs at locations of lower concentration chlorinated VOC’s.  While it is difficult or even impossible to discern the presence of VOC contamination based on the standard MIP log at this location, the LL MIP log exhibits robust signal to noise and the presence and location of contamination is easily defined. 

Collocated standard and LL MIP Logs displaying EC (shaded brown) and XSD (green)

Figure 8: Collocated standard and LL MIP Logs displaying EC (shaded brown) and XSD (green)

The MIP logs compared in Figure 9 again display closely spaced standard and LL MIP logs at locations of lower concentration chlorinated VOC’s.  The standard MIP log at this location is a bit easier to discern that contaminantes are likely present in this log however the LL MIP log displays strong signal to noise and the presence and location of contamination is easily defined.  

Collocated standard and LL MIP Logs displaying HPT Pressure (shaded blue) and XSD (green)

Figure 9: Collocated standard and LL MIP Logs displaying HPT Pressure (shaded blue) and XSD (green)

Specifications


FAQ's


Purchasing a System


LL MIP

Low Level Membrane Interface Probe

LL MIP logging in Denmark

LL MIP logging in Denmark

LL MIHPT Log

LL MIHPT Log

LL MIP Equipment

LL MIP Equipment

LL MIP logging in Denmark
LL MIHPT Log
LL MIP Equipment

Lower your MIP detection limits and increase your system usage with the LL MIP Pulse Flow controller

LL MIP was developed as a way to reduce MIP detection limits and offers the capability of reaching well below 100ppb total VOCs. These lowered detection limits means operators equipment sees more days in the field as plume investigations go to further extents and sites that MIP equipment would previously not have been used on are being investigated.

❯ Contact Us

LL MIP

Low Level Membrane Interface Probe


Lower your MIP detection limits and increase your system usage with the LL MIP Pulse Flow controller

LL MIP was developed as a way to reduce MIP detection limits and offers the capability of reaching well below 100ppb total VOCs. These lowered detection limits means operators equipment sees more days in the field as plume investigations go to further extents and sites that MIP equipment would previously not have been used on are being investigated.

❯ Contact Us

.

System Overview

Click on a section below to learn more about LL MIP.

LL MIP Pulse Flow Controller

Low Level MIP (LL MIP) is a technology developed by Geoprobe Systems® that greatly increases the sensitivity (and therefore utility) of the MIP logging tool.  The primary feature of LL MIP technology is that the carrier gas stream that sweeps the internal surface of the MIP membrane is pulsed.  This results in an increase in the concentration of VOC contaminant delivered to the MIP detectors.

Low Level MIP can be performed with any style of MIP probes or trunkline.  To perform this method an operator will need to add the MP9000 Pulse Flow Controller (Figure 1) to their current MIP system and be using a current version of DI Acquisition software.  The addition of the MP9000 to the system is simple and requires only the rearrangement of gas line connections.  This controller can then be easily removed from the system to return to standard MIP logging.  Switching between methods requires only a few minutes of time.

LL MIP Equipment: GC and detectors, FI6000, MP9000, MP6505

Figure 1: LL MIP Equipment: GC and detectors, FI6000, MP9000, MP6505

General Operation:
In standard MIP operation, the carrier gas continually sweeps across the membrane transporting contaminates to the detectors at the surface.  In the LL MIP method, the trunkline sweep flow is temporarily stopped when the MIP probe is brought to rest at a discrete depth in the soil.  Stopping the sweep gas flow allows the contaminant concentration to build behind the membrane.  This results in a larger and narrower contaminant response peak at the detectors (Fig. 2 and 3) for a given chemical concentration.  Switching valves located inside the MP9000 create separate flow paths for the MIP trunkline and detectors; trunkline flow can be stopped and restarted without impacting detector baseline or stability.   When the trunkline flow is restarted the contaminant mass (peak) is quickly swept to the surface with a trunkline flow rate of approximately 60ml/min. and is routed to the detectors via the MP9000.

Comparison of 0.5ppm Trichloroethylene response between standard MIP (50s-100s) and Low Level MIP (300s) methods

Figure 2: Comparison of 0.5ppm Trichloroethylene response between standard MIP (50s-100s) and Low Level MIP (300s) methods

The Low Level MIP operation is automated in DI Acquisition software version 3.0 or newer which controls the timed cycling of the MP9000 controller.  This software includes a subroutine for automatic triggering of LL MIP cycling at preset depth intervals during the MIP logging process.
The LL MIP method will greatly increase the sensitivity of a MIP system but the resulting detection limits are dependent on the sensitivity of the standard MIP system.  To achieve the lowest possible detection limits the probe and trunkline need to be new or verified clean with system blanks.  The detectors also need to be fully current within their maintenance program and sensitivity should be tested prior to mobilization.  Equipment that has been used to map high level contaminants will result in false positive results due to contaminant desorption from the membrane and return carrier gas line. 

Comparison of 0.2, 1.0, and 5.0ppm Trichloroethylene in standard and LL MIP methods

Figure 3: Comparison of 0.2, 1.0, and 5.0ppm Trichloroethylene in standard and LL MIP methods

The LL MIHPT Log in Figure 4 shows a highly permeable likely sandy formation in the top 22ft as indicated by the low EC and HPT pressure values.  Permeability stays fairly low throughout the log with occasional layers of lower permeability.  The predominant contaminant is seen from approximately 37ft to 51ft and is detected primarily on the PID and XSD.  Since the compounds respond on the XSD we know them to contain halogenated probably double bonded halogenated due to the PID confirmation, which to respond on the PID requires an ionization potential below the lamps 10.6eV.  The halogenated compounds that have lower ionization potential are typically double bonded chemicals.   The FID shows some responses within the first 5ft of the log which the other detectors do not confirm.  This is a common trait of natural gas or methane deposits which is non-halogenated and has a higher ionization potential than the lamp.  In the lower contaminant zone there is a confirmed response on the FID although lower in magnitude than the PID and XSD due to the lower sensitivity of the FID for these compounds.  Another item to note is that the PID displays consistent responses throughout the log, if it quite common for the PID to provide a background level response in LL MIP mode and the lower responses must be disregarded.

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, XSD, FID, and Est. K.

Figure 4: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, XSD, FID, and Est. K.

The LL MIHPT Log in Figure 5 shows a lowery permeability soils such as silts and clays in the top 45ft as indicated by the higher EC and HPT pressure values.  This site breaks into a sandy formation below 45ft indicated by the lower EC and HPT pressure below this point.  The predominant contaminant is seen from approximately 15ft to 33ft and is detected primarily on the PID and FID with no XSD response.  Since there is no response on the XSD we know it is a non-halogenated compound.  This log came from a former fuel service station so the PID and FID response likely is aromatic and nonaromatic fuel hydrocarbons.  

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

Figure 5: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

The LL MIHPT Log in Figure 6 shows a lower permeable clay at the top 5ft transitioning into likely a sandy silt formation down to 30ft as indicated by the low EC and HPT pressure values.  Below 30ft the formation is a fairly clean sand with low EC and HPT values. Permeability stays low throughout the log with occasional layers of lower permeability below 50ft.  There are different contaminants in this log with likely petroleum hydrocarbons in the top 30ft responding on the PID and FID only but then the XSD shows halogenated compounds 35ft-42ft and again from 52ft to the end of the boring.  These XSD responses are not confirmed on the PID which indicates that the ionization potential is likely too high for the PID to detect.  This typically occurs with single bonded halogenated compounds.  Confirmation groundwater samples were taken from these locations and 1,2-Dichloroethane was the predominant contaminant in the zone from 35ft to 42ft, in the lower section it was mostly Chloroform and Carbon Tetrachloride.   All of these compounds have ionization potentials over 11eV, so they would not be detected by the PID.   

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

Figure 6: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

Low level MIP will not replace standard MIP logging, but will expand MIP capabilities.  LL MIP is most useful when low contaminant concentrations are present.  Figure 7 displays an overlay of closely spaced standard MIP and LL MIP logs at locations of lower concentration petroleum and chlorinated VOC’s.  The LL MIP log exhibits robust signal to noise and the presence and location of contamination is easily defined.  The standard MIP log you might be able to discern there is a contaminant if you zoom in on the baseline but at times you can wonder if the basleine  movement is due to contaminant or is detector baseline drift.  This log displays petroleum hydrocarbon presence in the top half of the log with PID-FID responses and no XSD response and in the lower half of the log there are chlorinated contamoinants present as displayed by the PID-XSD responses.  Projects with contaminant concentrations too low to proceed with traditional MIP methods may be good candidates for investigation using the LL MIP method.  This technology will provide operators with the ability to track and map contaminant plumes down to concentrations at or below the 100ppb range for some contaminants.

Collocated standard (shaded green) and LL MIP Logs overlaid displaying (left to right) EC, MIP-PID, FID, and XSD

Figure 7: Collocated standard (shaded green) and LL MIP Logs overlaid displaying (left to right) EC, MIP-PID, FID, and XSD

The MIP logs compared in Figure 8 display closely spaced standard and LL MIP logs at locations of lower concentration chlorinated VOC’s.  While it is difficult or even impossible to discern the presence of VOC contamination based on the standard MIP log at this location, the LL MIP log exhibits robust signal to noise and the presence and location of contamination is easily defined. 

Collocated standard and LL MIP Logs displaying EC (shaded brown) and XSD (green)

Figure 8: Collocated standard and LL MIP Logs displaying EC (shaded brown) and XSD (green)

The MIP logs compared in Figure 9 again display closely spaced standard and LL MIP logs at locations of lower concentration chlorinated VOC’s.  The standard MIP log at this location is a bit easier to discern that contaminantes are likely present in this log however the LL MIP log displays strong signal to noise and the presence and location of contamination is easily defined.  

Collocated standard and LL MIP Logs displaying HPT Pressure (shaded blue) and XSD (green)

Figure 9: Collocated standard and LL MIP Logs displaying HPT Pressure (shaded blue) and XSD (green)

MIP Probe

System Overview


LL MIP Pulse Flow Controller

Low Level MIP (LL MIP) is a technology developed by Geoprobe Systems® that greatly increases the sensitivity (and therefore utility) of the MIP logging tool.  The primary feature of LL MIP technology is that the carrier gas stream that sweeps the internal surface of the MIP membrane is pulsed.  This results in an increase in the concentration of VOC contaminant delivered to the MIP detectors.

Low Level MIP can be performed with any style of MIP probes or trunkline.  To perform this method an operator will need to add the MP9000 Pulse Flow Controller (Figure 1) to their current MIP system and be using a current version of DI Acquisition software.  The addition of the MP9000 to the system is simple and requires only the rearrangement of gas line connections.  This controller can then be easily removed from the system to return to standard MIP logging.  Switching between methods requires only a few minutes of time.

LL MIP Equipment: GC and detectors, FI6000, MP9000, MP6505

Figure 1: LL MIP Equipment: GC and detectors, FI6000, MP9000, MP6505

General Operation:
In standard MIP operation, the carrier gas continually sweeps across the membrane transporting contaminates to the detectors at the surface.  In the LL MIP method, the trunkline sweep flow is temporarily stopped when the MIP probe is brought to rest at a discrete depth in the soil.  Stopping the sweep gas flow allows the contaminant concentration to build behind the membrane.  This results in a larger and narrower contaminant response peak at the detectors (Fig. 2 and 3) for a given chemical concentration.  Switching valves located inside the MP9000 create separate flow paths for the MIP trunkline and detectors; trunkline flow can be stopped and restarted without impacting detector baseline or stability.   When the trunkline flow is restarted the contaminant mass (peak) is quickly swept to the surface with a trunkline flow rate of approximately 60ml/min. and is routed to the detectors via the MP9000.

Comparison of 0.5ppm Trichloroethylene response between standard MIP (50s-100s) and Low Level MIP (300s) methods

Figure 2: Comparison of 0.5ppm Trichloroethylene response between standard MIP (50s-100s) and Low Level MIP (300s) methods

The Low Level MIP operation is automated in DI Acquisition software version 3.0 or newer which controls the timed cycling of the MP9000 controller.  This software includes a subroutine for automatic triggering of LL MIP cycling at preset depth intervals during the MIP logging process.
The LL MIP method will greatly increase the sensitivity of a MIP system but the resulting detection limits are dependent on the sensitivity of the standard MIP system.  To achieve the lowest possible detection limits the probe and trunkline need to be new or verified clean with system blanks.  The detectors also need to be fully current within their maintenance program and sensitivity should be tested prior to mobilization.  Equipment that has been used to map high level contaminants will result in false positive results due to contaminant desorption from the membrane and return carrier gas line. 

Comparison of 0.2, 1.0, and 5.0ppm Trichloroethylene in standard and LL MIP methods

Figure 3: Comparison of 0.2, 1.0, and 5.0ppm Trichloroethylene in standard and LL MIP methods

Example Logs


The LL MIHPT Log in Figure 4 shows a highly permeable likely sandy formation in the top 22ft as indicated by the low EC and HPT pressure values.  Permeability stays fairly low throughout the log with occasional layers of lower permeability.  The predominant contaminant is seen from approximately 37ft to 51ft and is detected primarily on the PID and XSD.  Since the compounds respond on the XSD we know them to contain halogenated probably double bonded halogenated due to the PID confirmation, which to respond on the PID requires an ionization potential below the lamps 10.6eV.  The halogenated compounds that have lower ionization potential are typically double bonded chemicals.   The FID shows some responses within the first 5ft of the log which the other detectors do not confirm.  This is a common trait of natural gas or methane deposits which is non-halogenated and has a higher ionization potential than the lamp.  In the lower contaminant zone there is a confirmed response on the FID although lower in magnitude than the PID and XSD due to the lower sensitivity of the FID for these compounds.  Another item to note is that the PID displays consistent responses throughout the log, if it quite common for the PID to provide a background level response in LL MIP mode and the lower responses must be disregarded.

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, XSD, FID, and Est. K.

Figure 4: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, XSD, FID, and Est. K.

The LL MIHPT Log in Figure 5 shows a lowery permeability soils such as silts and clays in the top 45ft as indicated by the higher EC and HPT pressure values.  This site breaks into a sandy formation below 45ft indicated by the lower EC and HPT pressure below this point.  The predominant contaminant is seen from approximately 15ft to 33ft and is detected primarily on the PID and FID with no XSD response.  Since there is no response on the XSD we know it is a non-halogenated compound.  This log came from a former fuel service station so the PID and FID response likely is aromatic and nonaromatic fuel hydrocarbons.  

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

Figure 5: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

The LL MIHPT Log in Figure 6 shows a lower permeable clay at the top 5ft transitioning into likely a sandy silt formation down to 30ft as indicated by the low EC and HPT pressure values.  Below 30ft the formation is a fairly clean sand with low EC and HPT values. Permeability stays low throughout the log with occasional layers of lower permeability below 50ft.  There are different contaminants in this log with likely petroleum hydrocarbons in the top 30ft responding on the PID and FID only but then the XSD shows halogenated compounds 35ft-42ft and again from 52ft to the end of the boring.  These XSD responses are not confirmed on the PID which indicates that the ionization potential is likely too high for the PID to detect.  This typically occurs with single bonded halogenated compounds.  Confirmation groundwater samples were taken from these locations and 1,2-Dichloroethane was the predominant contaminant in the zone from 35ft to 42ft, in the lower section it was mostly Chloroform and Carbon Tetrachloride.   All of these compounds have ionization potentials over 11eV, so they would not be detected by the PID.   

LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

Figure 6: LL MIHPT Log displaying graphs (left to right) EC, HPT Pressure (primary axis), Piezeometric profile (secondary axis), MIP-PID, FID, XSD, and Est. K.

MIP Method Comparison


Low level MIP will not replace standard MIP logging, but will expand MIP capabilities.  LL MIP is most useful when low contaminant concentrations are present.  Figure 7 displays an overlay of closely spaced standard MIP and LL MIP logs at locations of lower concentration petroleum and chlorinated VOC’s.  The LL MIP log exhibits robust signal to noise and the presence and location of contamination is easily defined.  The standard MIP log you might be able to discern there is a contaminant if you zoom in on the baseline but at times you can wonder if the basleine  movement is due to contaminant or is detector baseline drift.  This log displays petroleum hydrocarbon presence in the top half of the log with PID-FID responses and no XSD response and in the lower half of the log there are chlorinated contamoinants present as displayed by the PID-XSD responses.  Projects with contaminant concentrations too low to proceed with traditional MIP methods may be good candidates for investigation using the LL MIP method.  This technology will provide operators with the ability to track and map contaminant plumes down to concentrations at or below the 100ppb range for some contaminants.

Collocated standard (shaded green) and LL MIP Logs overlaid displaying (left to right) EC, MIP-PID, FID, and XSD

Figure 7: Collocated standard (shaded green) and LL MIP Logs overlaid displaying (left to right) EC, MIP-PID, FID, and XSD

The MIP logs compared in Figure 8 display closely spaced standard and LL MIP logs at locations of lower concentration chlorinated VOC’s.  While it is difficult or even impossible to discern the presence of VOC contamination based on the standard MIP log at this location, the LL MIP log exhibits robust signal to noise and the presence and location of contamination is easily defined. 

Collocated standard and LL MIP Logs displaying EC (shaded brown) and XSD (green)

Figure 8: Collocated standard and LL MIP Logs displaying EC (shaded brown) and XSD (green)

The MIP logs compared in Figure 9 again display closely spaced standard and LL MIP logs at locations of lower concentration chlorinated VOC’s.  The standard MIP log at this location is a bit easier to discern that contaminantes are likely present in this log however the LL MIP log displays strong signal to noise and the presence and location of contamination is easily defined.  

Collocated standard and LL MIP Logs displaying HPT Pressure (shaded blue) and XSD (green)

Figure 9: Collocated standard and LL MIP Logs displaying HPT Pressure (shaded blue) and XSD (green)





Videos



Resources

Click on a section below to view information.



LL MIP Setup

Literature Type: Technical Documents
Download File: File LL MIP Setup.pdf





LL MIP E-Specs

Literature Type: E-Specs Technical Documents
Download File: File ps_2013_di_ll-mip_0.pdf





LL MIP Standard Operating Procedure

Literature Type: SOPs Technical Documents
Download File: File LL MIP V.2 SOP_0.pdf





LL MIP Overview

Literature Type: Technical Documents
Download File: File LL MIP Overview_0.pdf





MTBE using LL MIP Method

Literature Type: Technical Documents
Download File: File MTBE using LL MIP Method.pdf



LL MIP logging in Denmark

LL MIP logging in Denmark

LL MIHPT Log

LL MIHPT Log

LL MIP Equipment

LL MIP Equipment

© 2018 Kejr, Inc. | Privacy Policy