To summarize, the results may be interpreted through the lens of the critical performance characteristics that are sought for consolidants of earthen surface finishes for this site. Because of the reattachment treatment that must also be performed at Mesa Verde, we need to consider not only the effectiveness of the consolidants in imparting increased cohesive strength, but also the resulting effects on hygric and hydric behavior.
Thus, the improvement of cohesive strength is only a basic required performance characteristic, a prerequisite before a consolidant can be considered further. The Silbond, Funcosil SAE and Antihygro treatments all increased surface cohesion within a similar range, all appropriate results, while the loss of surface cohesion seen with the Conservare is a poor indication of its utility. The water-treated samples surprisingly also showed an increase in surface cohesion after wet-dry cycling, although not as great as the consolidated samples.
Of great importance is the continued plasticity after treatment so that future reattachment may be performed. Since plasticity has been shown to be related to hygric expansion, the differential transducer assessments of hygric dimensional change may be considered an indication of plasticity. Silbond had the least decrease in dimensional change as compared to the control so is considered the most appropriate in terms of this property.
In opposition to this, dimensional stability is another critical performance characteristic. When the tests for hydric change are interpreted with only this in mind, the water yielded the most appropriate results. This requires a decision as to which characteristic is more important: continued plasticity or dimensional stability.
In this case, for the reattachment to be possible, plasticity takes precedence, making the results for Silbond the most appropriate.
Dimensional stability was also measured on bulk samples by wet-dry cycling, in which the most stabilizing treatments were seen to be Silbond and Funcosil Antihygro with SAE.
We would ideally see no adverse change in water permeability in both liquid and vapor form after consolidation. This characteristic was measured by microdrop absorption and a loss in liquid water permeability was seen for all chemically treated samples. While water repellency can be a protective quality, it limits our ability to perform the reattachment treatment since the finishes need to be wet to activate their plasticity and allow them to absorb the gelatin solution.
Thus, in this case, absorption is a desired quality and Conservare has the most appropriate results due to its moderate water repellency.
Lastly, it is critical that consolidation produce no adverse change in the color or texture of the finishes. This was observed to be the least for the water treated samples, as would be expected. And of the chemically treated samples, the Silbond produced the least color change.
Texture remained relatively unchanged as a result of consolidation, with the exception of the Antihygro treatment in which an unacceptable level of alteration was seen. Since there were a few findings that are inappropriate for this site, consolidant selection now becomes a game of elimination with the darkest shaded cells representing inappropriate findings. Reading horizontally, the water treatment had three sets very good results, but was very unstable during wet-dry cycling. Conservare decreased the surface cohesion after wet-dry cycling, and therefore cannot be considered as a consolidant for these finishes. Silbond performed very well for all categories except microdrop absorption which is an essential performance characteristic and renders the Silbond inappropriate. Antihygro-treated samples performed fairly well but the loss of texture it produced is not acceptable. The Funcosil SAE 300E without pretreatment appears to be the best option with moderate hygric dimensional change, increased stability under wet-dry cycling, moderate water repellency, greatly increased surface cohesion, minimal color change, and very little surface effects after treatment.

During wet-dry cycling the surface effects on each type of treated sample varied greatly. Water was absorbed rapidly into the untreated and water-treated samples which then deformed by cupping. Absorption was much slower for the consolidated samples, the Silbond-treated samples having the slowest absorption rate. Moisture transmission seemed be slowed for other consolidated samples which retained a dark, wet appearance far longer than the controls.

The effect of consolidation on sample surfaces was recorded for each of the four chemical consolidants, all of which remained relatively unchanged with the exception of the Antihygro-treated samples. These seemed to expand and retain that position after curing.

From left to right, samples were treated with Antihygro and SAE, Funcosil SAE300E alone, Conservare OH100 and Silbond 40.

Cracks in the finish disappeared and the surface took on a mottled texture after treatment.

From left to right, samples were treated with Antihygro and SAE, Funcosil SAE300E alone, Conservare OH100 and Silbond 40.

Surface cohesion was evaluated through a test in which material is removed from the samples with clear tape and photographed. The photos are then processed digitally to analyze the relative density of removed material for each treatment type.

Results before wet-dry cycling showed minimal variation from the control. The Conservare and Funcosil consolidant increased surface cohesion while water, Silbond and Antihygro unexpectedly decreased surface cohesion. This decrease for Silbond-treated samples may be attributed to its slow curing time, while the results for the Antihygro-treated samples may be due to a change in surface texture. After wet-dry cycling the variation in results was much more pronounced with the Silbond and both Funcosil-treated samples showing great improvements in surface cohesion. The Conservare-treated samples however, greatly decreased in surface cohesion, a result that when paired with the wet-dry cycling results for Phase I, indicates that earthen surface finishes treated with this consolidant may be particularly sensitive to exposure to liquid water.

Before treatment, the samples had considerable cracking due to the production process and sample composition.

In transfer to the metal sheet for wet-dry cycling, the cracking was exacerbated, especially in the control and water-treated samples. In the chemically-treated samples, the consolidant appeared to act with slight adhesive qualities to maintain bulk cohesion. The color of the treated samples was slightly darker than the untreated samples after consolidation and before wet-dry cycling. Wet-dry cycling was performed by spraying the samples, allowing the water to absorb and evaporate for 30 minutes, and placing the samples in an oven at 60°C for 1 – 2 hours. 40 cycles were performed over a two-week period.

After wet-dry cycling it is clear that the control and water-treated samples have a much greater hydric response than the chemically consolidated samples and exhibit significant planar deformation. Of the chemically consolidated samples, Silbond and Antihygro elicited the least response to cycling and Conservare showed the greatest response. Darkening of the samples due to consolidation seems to have been moderated by the wet-dry cycling, with the Silbond-treated samples having the lightest value.

Evaluation for hygric dimensional change was performed by subjecting samples to varying relative humidity and measuring corresponding dimensional fluctuations.

The relative humidity was varied as seen in the yellow line and was consistent for all twelve samples tested. The bright blue line marks the corresponding expansion and contraction of the sample. The noise in this data was minimized by calculating a running average with a period of ten, as represented by the light blue line.  The total dimensional change for each period was then able to be measured and averaged for each sample.

Two samples were tested of each treatment type so both figures were averaged to produce a number that is an indicator of the expansion and contraction for each treatment type. The results for each sample are seen here in gray and the average appears in red. Samples treated with water showed the greatest increase in dimensional stability, however all treated samples gained stability as well. Of these, Antihygro maintained the most stability and Silbond produced the least significant gains.