Determining The Asphalt Binder Content of Hot Mix Asphalt (Hma) by The Ignition Method Fop For Aashto T 308 Scope

Download as pdf or txt
Download as pdf or txt
You are on page 1of 14

ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

DETERMINING THE ASPHALT BINDER CONTENT OF HOT MIX ASPHALT


(HMA) BY THE IGNITION METHOD
FOP FOR AASHTO T 308

Scope

This procedure covers the determination of asphalt binder content of hot mix asphalt (HMA)
by ignition of the binder in accordance with AASHTO T 308-10.
Overview

The sample is heated in a furnace at 538°C (1000°F) or less; samples may be heated by
convection or direct infrared irradiation (IR). The aggregate remaining after burning can be
used for sieve analysis using the FOP for AASHTO T 30.

Some agencies allow the use of recycled HMA. When using recycled HMA, check with the
agency for specific correction procedures.

Binder in the HMA is ignited in a furnace. Asphalt binder content is calculated as the
difference between the initial mass of the HMA and the mass of the residual aggregate,
correction factor, and moisture content. The asphalt binder content is expressed as percent of
moisture-free mix mass.

Two methods, A and B, are presented.

Apparatus
Note 1: The apparatus must be calibrated for the specific mix design. See “Correction Factors” at the end of
this FOP.
There are two methods – A and B. The apparatus for the two methods are the same except
that the furnace for Method A has an internal balance.

 Ignition Furnace: A forced-air ignition furnace that heats the specimens by either the
convection or direct IR irradiation method. The convection-type furnace must be capable
of maintaining the temperature at 578°C (1072°F).

For Method A, the furnace will be equipped with an internal scale thermally isolated
from the furnace chamber and accurate to 0.1 g. The scale shall be capable of
determining the mass of a 3500 g sample in addition to the sample baskets. A data
collection system will be included so that mass can be automatically determined and
displayed during the test. The furnace shall have a built-in computer program to
calculate the change in mass of the sample baskets and provide for the input of a
correction factor for aggregate loss. The furnace shall provide a printed ticket with the
initial specimen mass, specimen mass loss, temperature compensation, correction factor,
corrected binder content, test time, and test temperature. The furnace shall provide an
audible alarm and indicator light when the sample mass loss does not exceed 0.01 percent
of the total sample mass for three consecutive minutes.

T308_short_12.docx Asphalt 14 - 1 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

Note 2: The furnace shall be designed to permit the operator to change the ending mass loss percentage
from 0.01 percent to 0.02 percent.
For both Method A and Method B, the furnace chamber dimensions shall be adequate to
accommodate a 3500 g sample. The furnace door shall be equipped so that it cannot be
opened during the ignition test. A method for reducing furnace emissions shall be
provided and the furnace shall be vented so that no emissions escape into the laboratory.
The furnace shall have a fan to pull air through the furnace to expedite the test and to
eliminate the escape of smoke into the laboratory.

 Sample Basket Assembly: consisting of sample basket(s), catch pan, and basket guards.
Sample basket(s) will be of appropriate size allowing samples to be thinly spread and
allowing air to flow through and around the sample particles. Sets of two or more
baskets shall be nested. A catch pan: of sufficient size to hold the sample basket(s) so
that aggregate particles and melting binder falling through the screen mesh are caught.
Basket guards will completely enclose the basket and be made of screen mesh, perforated
stainless steel plate, or other suitable material.

 Thermometer, or other temperature measuring device, with a temperature range of 10 -


260°C (50-500°F).

 Oven capable of maintaining 110 ±5°C (230 ±9°F).

 Balance or scale: Capacity sufficient for the sample mass and conforming to the
requirements of M 231, Class G2.

 Safety equipment: Safety glasses or face shield, high temperature gloves, long sleeved
jacket, a heat resistant surface capable of withstanding 650°C (1202°F), a protective cage
capable of surrounding the sample baskets during the cooling period, and a particle mask
for use during removal of the sample from the basket assembly.

 Miscellaneous equipment: A pan larger than the sample basket(s) for transferring sample
after ignition, spatulas, bowls, and wire brushes.

Sampling

1. Obtain samples of HMA in accordance with the FOP for AASHTO T 168.

2. Reduce HMA samples in accordance with the FOP for AASHTO R 47.

3. If the mixture is not sufficiently soft to separate with a spatula or trowel, place it in a
large flat pan in an oven at 110 ±5°C (230 ±9°F) until soft enough.

4. Test sample size shall conform to the mass requirement shown in Table 1.
Note 3: When the mass of the test specimen exceeds the capacity of the equipment used or for large
samples of fine mixes, the test specimen may be divided into suitable increments, tested, and the results
appropriately combined through a weighted average for calculation of the binder content.

T308_short_12.docx Asphalt 14 - 2 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

Table 1
Nominal Maximum Minimum Maximum
Aggregate Size* Mass Specimen Mass Specimen
mm (in.) g g
37.5 (1 ½) 4000 4500
25.0 (1) 3000 3500
19.0 (3/4) 2000 2500
12.5 (1/2) 1500 2000
9.5 (3/8) 1200 1700
4.75 (No. 4) 1200 1700

* One sieve larger than the first sieve to retain more than 10 percent of the material using an agency
specified set of sieves based on cumulative percent retained. Where large gaps in specification sieves exist,
intermediate sieve(s) may be inserted to determine nominal maximum size.

Procedure – Method A (Internal Balance)

1. For the convection-type furnace, preheat the ignition furnace to 538°C (1000°F) or to the
temperature determined in the “Correction Factor” section, Step 9 of this method.
Manually record the furnace temperature (set point) prior to the initiation of the test if the
furnace does not record automatically. For the direct IR irradiation-type furnace, use the
same burn profile as used during the correction factor determination.

2. Dry the sample to constant mass, according to the FOP for AASHTO T 329; or determine
the moisture content of a companion sample in accordance with the FOP for AASHTO
T 329.

3. Determine and record the mass to the nearest 0.1 g of the sample basket assembly.

4. Evenly distribute the sample in the sample basket assembly, taking care to keep the
material away from the edges of the basket. Use a spatula or trowel to level the sample.

5. Determine and record the total mass of the sample and sample basket assembly to the
nearest 0.1 g. Calculate and record the initial mass of the sample (total mass minus the
mass of the sample basket assembly) to the nearest 0.1 g. Designate this mass as (Mi).

6. Record the correction factor or input into the furnace controller for the specific HMA.

7. Input the initial mass of the sample (Mi) into the ignition furnace controller. Verify that
the correct mass has been entered.
CAUTION: Operator should wear safety equipment – high temperature gloves, face shield, fire-retardant
shop coat – when opening the door to load or unload the sample.
8. Open the chamber door and gently set the sample basket assembly in the furnace.
Carefully position the sample basket assembly so it is not in contact with the furnace

T308_short_12.docx Asphalt 14 - 3 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

wall. Close the chamber door and verify that the sample mass displayed on the furnace
scale equals the total mass of the sample and sample basket assembly recorded in Step 5
within ±5 g.
Note 4: Furnace temperature will drop below the set point when the door is opened, but will recover when
the door is closed and ignition begins. Sample ignition typically increases the temperature well above the
set point – relative to sample size and binder content.
9. Initiate the test by pressing the start button. This will lock the sample chamber and start
the combustion blower.
Safety note: Do not attempt to open the furnace door until the asphalt binder has been completely
burned off.
10. Allow the test to continue until the stable light and audible stable indicator indicate that
the change in mass does not exceed 0.01 percent for three consecutive minutes. Press the
stop button. This will unlock the sample chamber and cause the printer to print out the
test results.
Note 5: An ending mass loss percentage of 0.02 may be used, if allowed by the agency, when aggregate
that exhibits an excessive amount of loss during ignition testing is used.
11. Open the chamber door, remove the sample basket assembly, and place on the cooling
plate or block. Place the protective cage over the sample basket assembly and allow it to
cool to room temperature (approximately 30 minutes).

12. Determine and record the total after ignition mass to the nearest 0.1 g. Calculate and
record the mass of the sample, after ignition (total after ignition mass minus the mass of
the sample basket assembly) to the nearest 0.1 g. Designate this mass as Mf.

13. Use the asphalt binder content percentage from the printed ticket. Subtract the moisture
content from the printed ticket asphalt binder content and report the difference as the
corrected asphalt binder content.

Pb = BC – MC – Cf (if not input in the furnace controller)

where:
Pb = the corrected asphalt binder content as a percent by mass of the HMA

BC = asphalt binder content shown on printed ticket

MC = moisture content of the companion HMA sample, percent, as


determined by the FOP for AASHTO T 329 (if the specimen was oven-dried
prior to initiating the procedure, MC=0)

Cf = correction factor as a percent by mass of the HMA sample

14. Asphalt binder content percentage can also be calculated using the formula from “Method
B” Step 16.

T308_short_12.docx Asphalt 14 - 4 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

Procedure – Method B (External Balance)

1. Preheat the ignition furnace to 538°C (1000°F) or to the temperature determined in the
“Correction Factor” section, Step 9 of this method. Manually record the furnace
temperature (set point) prior to the initiation of the test if the furnace does not record
automatically.

2. Dry the sample to constant mass, according to the FOP for AASHTO T 329; or determine
the moisture content of a companion sample in accordance with the FOP for AASHTO
T 329.

3. Determine and record the mass of the sample basket assembly to the nearest 0.1 g.

4. Place the sample basket(s) in the catch pan. Evenly distribute the sample in the sample
basket(s), taking care to keep the material away from the edges of the basket. Use a
spatula or trowel to level the sample.

5. Determine and record the total mass of the sample and sample basket assembly to the
nearest 0.1 g. Calculate and record the initial mass of the sample (total mass minus the
mass of the sample basket assembly) to the nearest 0.1 g. Designate this mass as (Mi).

6. Record the correction factor for the specific HMA.

7. Open the chamber door and gently set the sample basket assembly in the furnace.
Carefully position the sample basket assembly so it is not in contact with the furnace
wall. Burn the HMA sample in the furnace for 45 minutes or the length of time
determined in the “Correction Factors” section.

8. Open the chamber door, remove the sample basket assembly, and place on the cooling
plate or block. Place the protective cage over the sample and allow it to cool to room
temperature (approximately 30 min).

9. Determine and record the total after ignition mass to the nearest 0.1 g. Calculate and
record the mass of the sample, after ignition (total after ignition mass minus the mass of
the sample basket assembly) to the nearest 0.1 g.

10. Place the sample basket assembly back into the furnace.

11. Burn the sample for at least 15 minutes after the furnace reaches the set temperature.

12. Open the chamber door, remove the sample basket assembly, and place on the cooling
plate or block. Place the protective cage over the sample basket assembly and allow it to
cool to room temperature (approximately 30 min.).

T308_short_12.docx Asphalt 14 - 5 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

13. Determine and record the total after ignition mass to the nearest 0.1 g. Calculate and
record the mass of the sample, after ignition (total after ignition mass minus the mass of
the sample basket assembly) to the nearest 0.1 g.

14. Repeat Steps 10 through 13 until the change in measured mass of the sample after
ignition does not exceed 0.01 percent of the previous sample mass after ignition.
Note56: An ending mass loss percentage of 0.02 may be used, if allowed by the agency, when aggregate
that exhibits an excessive amount of loss during ignition testing is used.
15. Determine and record the total after ignition mass to the nearest 0.1 g. Calculate and
record the mass of the sample, after ignition (total after ignition mass minus the mass of
the sample basket assembly) to the nearest 0.1 g. Designate this mass as Mf.

16. Calculate the asphalt binder content of the sample as follows:

100
where:
Pb = the corrected asphalt binder content as a percent by mass of the HMA
sample

Mf = the final mass of aggregate remaining after ignition

Mi = the initial mass of the HMA sample prior to ignition

Cf = correction factor as a percent by mass of the HMA sample

MC= moisture content of the companion HMA sample, percent, as


determined by the FOP for AASHTO T 329 (if the specimen was
oven-dried prior to initiating the procedure, MC=0).

Example

Correction Factor = 0.42

Moisture Content = 0.04

Initial Mass of Sample and Basket = 5292.7

Mass of Basket Assembly = 2931.5

Mi = 2361.2

Total Mass after First ignition + basket = 5154.4

Sample Mass after First ignition = 2222.9

T308_short_12.docx Asphalt 14 - 6 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

Sample Mass after additional 15 min ignition = 2222.7

2222.9 2222.7
100 0.009
2222.9

Not greater than 0.01 percent, so Mf = 2222.7

2361.2 2222.7
100 0.42 0.04 5.41%
2361.2

Pb= 5.41%

Gradation

1. Empty contents of the basket(s) into a flat pan, being careful to capture all material. Use
a small wire brush to ensure all residual fines are removed from the baskets.
Note 6: Particle masks are a recommended safety precaution.
2. Perform the gradation analysis in accordance with the FOP for AASHTO T 30.

Report

 Results on forms approved by the agency

 Method of test (A or B)

 Corrected asphalt binder content, Pb, per agency standard

 Correction factor, Cf , to 0.01 percent

 Temperature compensation factor (if applicable)

 Total percent loss

 Sample mass

 Moisture content to 0.01%

 Test temperature

Attach the original printed ticket with all intermediate values (continuous tape) to the report
for furnaces with internal balances.

T308_short_12.docx Asphalt 14 - 7 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

Correction Factors

Asphalt Binder and Aggregate

Asphalt binder content results may be affected by the type of aggregate in the mixture and by
the ignition furnace. Asphalt binder and aggregate correction factors must, therefore, be
established by testing a set of correction specimens for each Job Mix Formula (JMF) mix
design. Each ignition furnace will have its own unique correction factor determined in the
location where testing will be performed.
This procedure must be performed before any acceptance testing is completed, and repeated
each time there is a change in the mix ingredients or design. Any changes greater than 5
percent in stockpiled aggregate proportions should require a new correction factor.
Historical data or scientific studies may be used to determine the correction factor(s) in lieu
of using this testing procedure if the testing agency provides reference to the studies/data.
All correction samples will be prepared by a central / regional laboratory unless otherwise
directed.

Asphalt binder correction factor: A correction factor must be established by testing a set of
correction specimens for each Job Mix Formula (JMF). Certain aggregate types may result in
unusually high correction factors (> 1.00 percent). Such mixes should be corrected and
tested at a lower temperature as described below.

Aggregate correction factor: Due to potential aggregate breakdown during the ignition
process, a correction factor will need to be determined for the following conditions:

a. Aggregates that have a proven history of excessive breakdown

b. Aggregate from an unknown source.

This correction factor will be used to adjust the acceptance gradation test results obtained
according to the FOP for AASHTO T 30.

Procedure

1. Obtain samples of aggregate in accordance with the FOP for AASHTO T 2.

2. Obtain samples of asphalt binder in accordance with the FOP for AASHTO T 40.
Note 7: Include other additives that may be required by the JMF.
3. Prepare an initial, or “butter,” mix at the design asphalt binder content. Mix and discard
the butter mix prior to mixing any of the correction specimens to ensure accurate asphalt
content.

4. Prepare two correction specimens at the JMF design asphalt binder content. Aggregate
used for correction specimens shall be sampled from material designated for use on the
project. An agency approved method will be used to combine aggregate. An additional
“blank” specimen shall be batched and tested for aggregate gradation in accordance with

T308_short_12.docx Asphalt 14 - 8 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

the FOP for AASHTO T 30. The gradation from the “blank” shall fall within the agency
specified mix design tolerances.

5. Place the freshly mixed specimens directly into the sample basket assembly. If mixed
specimens are allowed to cool prior to placement in the sample basket assembly, the
specimens must be dried to constant mass according to the FOP for AASHTO T 329. Do
not preheat the sample basket assembly.

6. Test the specimens in accordance with Method A or Method B of the procedure.

7. Once both of the correction specimens have been burned, determine the asphalt binder
content for each specimen by calculation or from the printed oven tickets, if available.

8. If the difference between the asphalt binder contents of the two specimens exceeds 0.15
percent, repeat with two more specimens and, from the four results, discard the high and
low result. Determine the correction factor from the two original or remaining results, as
appropriate. Calculate the difference between the actual and measured asphalt binder
contents for each specimen to 0.01 percent. The asphalt binder correction factor, Cf, is
the average of the differences expressed as a percent by mass of HMA.

9. If the asphalt binder correction factor exceeds 1.00 percent, the test temperature must be
lowered to 482 ±5°C (900 ±8°F) and new samples must be burned. The temperature for
determining the asphalt binder content of HMA samples by this procedure shall be the
same temperature determined for the correction samples.

10. For the direct IR irradiation-type burn furnaces, the default burn profile should be used
for most materials. The operator may select burn-profile Option 1 or Option 2 to optimize
the burn cycle. The burn profile for testing HMA samples shall be the same burn profile
selected for correction samples.

Option 1 is designed for aggregate that requires a large asphalt binder correction
factor (greater than 1.00 percent) – typically very soft aggregate (such as dolomite).

Option 2 is designed for samples that may not burn completely using the default
burn profile.

11. Perform a gradation analysis on the residual aggregate in accordance with the FOP for
AASHTO T 30, if required. The results will be utilized in developing an “Aggregate
Correction Factor” and should be calculated and reported to 0.1 percent.

12. From the gradation results subtract the percent passing for each sieve, for each sample,
from the percent passing each sieve of the “Blank” specimen gradation results from
Step 4.

13. Determine the average difference of the two values. If the difference for any single sieve
exceeds the allowable difference of that sieve as listed in Table 2, then aggregate

T308_short_12.docx Asphalt 14 - 9 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

gradation correction factors (equal to the resultant average differences) for all sieves shall
be applied to all acceptance gradation test results determined by the FOP for AASHTO
T 30. If the 75 µm (No. 200) is the only sieve outside the limits in Table 2, apply the
aggregate correction factor to only the 75 µm (No. 200) sieve.

Table 2
Permitted Sieving Difference
Sieve Allowable Difference
Sizes larger than or equal to 2.36 mm (No.8) ± 5.0%
Sizes larger than to 75 µm (No.200) and smaller than 2.36 ± 3.0%
mm (No.8)
Sizes 75 µm (No.200) and smaller ± 0.5%

Examples:

Correction Correction Correction Difference Avg. Sieves to


Factor Factor Factor 1/2 Diff. adjust
Sieve Size Blank Sample Sample #1 Sample #2
mm (in.) % Passing % Passing % Passing
19.0 (3/4) 100 100 100 0/0 0.0
12.5 (1/2) 86.3 87.4 86.4 -1.1/-0.1 -0.6
9.5 (3/8) 77.4 76.5 78.8 +0.9/-1.4 -0.2
4.75 (No. 4) 51.5 53.6 55.9 -2.1/-4.4 -3.2
2.36 (No. 8) 34.7 36.1 37.2 -1.4/-2.5 -2.0
01.18 (No. 16) 23.3 25.0 23.9 -1.7/-0.6 -1.2
0.600 (No. 30) 16.4 19.2 18.1 -2.8/-1.7 -2.2
0.300 (No. 50) 12.0 11.1 12.7 +0.9/-0.7 +0.1
0.150 (No. 100) 8.1 9.9 6.3 -1.8/+1.8 0.0
75 µm (No. 200) 5.5 5.9 6.2 -0.4/-0.7 -0.6 - 0.6

In this example all acceptance gradation test results (FOP for AASHTO T 30) performed on
the residual aggregate would have an “Aggregate Correction Factor”. This factor would be
- 0.6% on the 75 µm (No. 200) sieve and would be applied to the percent passing 75 µm
(No.200) sieve.

Correction Correction Correction Difference Avg. Sieves to


Factor Factor Factor 1/2 Diff. adjust
Sieve Size Blank Sample Sample #1 Sample #2
mm (in.) % Passing % Passing % Passing
19.0 (3/4) 100 100 100 0/0 0.0 0.0
12.5 (1/2) 86.3 87.4 86.4 -1.1/-0.1 -0.6 -0.6
9.5 (3/8) 77.4 76.5 78.8 +0.9/-1.4 -0.2 -0.2
4.75 (No. 4) 51.5 55.6 57.9 -4.1/-6.4 -5.2 -5.2
2.36 (No. 8) 34.7 36.1 37.2 -1.4/-2.5 -2.0 -2.0
01.18 (No. 16) 23.3 25.0 23.9 -1.7/-0.6 -1.2 -1.2
0.600 (No. 30) 16.4 19.2 18.1 -2.8/-1.7 -2.2 -2.2
0.300 (No. 50) 12.0 11.1 12.7 +0.9/-0.7 +0.1 +0.1
0.150 (No. 100) 8.1 9.9 6.3 -1.8/+1.8 0.0 0.0
75 µm (No. 200) 5.5 5.9 6.2 -0.4/-0.7 -0.6 -0.6

T308_short_12.docx Asphalt 14 - 10 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

In this example all acceptance gradation test results (FOP for AASHTO T 30) performed on
the residual aggregate would have an “Aggregate Correction Factor”. The correction factor
for each sieve must be applied because the average difference on the 4.75mm (No. 4) is
outside the tolerance from Table 2.

T308_short_12.docx Asphalt 14 - 11 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (12)

T308_short_12.docx Asphalt 14 - 12 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (09)

PERFORMANCE EXAM CHECKLIST

DETERMINING THE ASPHALT BINDER CONTENT OF HOT MIX ASPHALT


(HMA) BY THE IGNITION METHOD
FOP FOR AASHTO T 308

Participant Name ______________________________ Exam Date ______________


Record the symbols “P” for passing or “F” for failing on each step of the checklist.

Procedure Element Trial 1 Trial 2


1. Oven at correct temperature 538°C (1000°F) or correction factor temperature? _____ _____
Or: for IR ovens, correct burn profile applied?
2. Sample reduced to correct size? _____ _____
3. HMA sample or companion moisture sample taken and dried per
FOP for AASHTO T 329? _____ _____
4. Mass of sample basket assembly recorded to 0.1 g? _____ _____
5. With pan below basket(s) sample evenly distributed in basket(s)? _____ _____
6. Sample conforms to the required mass and mass recorded to 0.1 g? _____ _____

7. Method A
a. Initial mass entered into furnace controller? _____ _____
b. Sample correctly placed into furnace? _____ _____
c. Test continued until stable indicator signals? _____ _____
d. Uncorrected binder content obtained on printed ticket? _____ _____
e. Sample mass determined to nearest 0.1 g.? _____ _____

8. Method B
a. Sample correctly placed into furnace? _____ _____
b. Sample burned for 45 min or time determined by correction process? _____ _____
c. Sample cooled to room temperature? _____ _____
d. Sample burned to constant mass? _____ _____
e. Sample mass determined to nearest 0.1 g.? _____ _____
f. Uncorrected binder content calculated correctly and recorded? _____ _____

OVER

T308_pr_09.docx Asphalt 6-19 Pub. October 2012


ASPHALT WAQTC / IDAHO FOP AASHTO T 308 (09)

Procedure Element Trial 1 Trial 2

9. Binder content corrected for Correction Factor if needed? _____ _____


10. Binder content corrected for moisture per T 329 if needed? _____ _____
11. Corrected binder content recorded? _____ _____
12. Contents of the basket(s) carefully emptied into a pan? _____ _____

Comments: First attempt: Pass Fail Second attempt: Pass Fail

Examiner Signature _______________________________WAQTC #:_______________

T308_pr_09.docx Asphalt 6-20 Pub. October 2012

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy